INDEX

Explanations

phrases related to specific locations, particularly those ending with "wood"

oai_token-act-pair · gpt-3.5-turbo

the word "wood" in various 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

 polar

-0.79

ional

-0.71

Magikarp

-0.67

oresc

-0.67

 emanc

-0.66

 dishon

-0.63

olar

-0.63

 delinquent

-0.63

 superhuman

-0.62

 federally

-0.62

POSITIVE LOGITS

mere

1.02

wood

1.01

vale

0.96

stock

0.92

leigh

0.92

 Springs

0.90

working

0.88

hurst

0.86

seed

0.85

 Preferred

0.84

Activations Density 0.013%

phrases related to specific locations, particularly those ending with "wood"

the word "wood" in various contexts

No Comments

No Known Activations