INDEX

Explanations

words related to something being thoroughly searched, investigated, or examined

oai_token-act-pair · gpt-3.5-turbo

words and variations related to scurrying or hurried movements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Adds

-0.61

 Samar

-0.56

 nort

-0.56

bod

-0.53

 subt

-0.53

ndra

-0.53

 acet

-0.52

 thirst

-0.52

 craft

-0.51

 envy

-0.50

POSITIVE LOGITS

arios

0.95

itars

0.78

tle

0.76

riage

0.75

abba

0.75

inav

0.74

otto

0.72

riages

0.72

olded

0.71

uli

0.71

Activations Density 0.062%

words related to something being thoroughly searched, investigated, or examined

words and variations related to scurrying or hurried movements

No Comments

No Known Activations