INDEX

Explanations

phrases relating to comparisons and disparities within different groups or entities

oai_token-act-pair · gpt-3.5-turbo

the phrase "within" indicating boundaries or limitations 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 @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

mist

-0.78

ãĤ·ãĥ£

-0.69

kaya

-0.67

ammy

-0.66

ador

-0.66

ãĥĵ

-0.66

vous

-0.66

disk

-0.65

rog

-0.64

intend

-0.64

POSITIVE LOGITS

 parentheses

0.80

 confines

0.73

 bounds

0.71

isine

0.70

 brackets

0.69

imore

0.68

groups

0.68

 academia

0.68

Ĥª

0.67

 Skydragon

0.66

Activations Density 0.034%

phrases relating to comparisons and disparities within different groups or entities

the phrase "within" indicating boundaries or limitations in various contexts

No Comments

No Known Activations