INDEX

Explanations

mentions of the term "Dep" followed by a number (e.g., Dep 9, Dep 10) - potentially related to a specific department or designation

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "Dep" and its variations in different contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Pose

-0.91

 Hole

-0.82

nings

-0.81

SPONSORED

-0.79

MER

-0.78

 deaf

-0.76

enegger

-0.76

FORE

-0.76

çķ

-0.76

è¯

-0.75

POSITIVE LOGITS

recated

1.18

artments

1.12

uty

1.09

utation

1.08

ository

1.07

osit

1.05

artment

1.03

orters

1.02

leted

1.01

ressed

0.98

Activations Density 0.013%

mentions of the term "Dep" followed by a number (e.g., Dep 9, Dep 10) - potentially related to a specific department or designation

occurrences of the term "Dep" and its variations in different contexts

No Comments

No Known Activations