INDEX

Explanations

numbers representing ratings or votes

oai_token-act-pair · gpt-3.5-turbo

repeated numeric values, specifically voting tallies

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

xual

-0.75

ĸļ

-0.73

ean

-0.71

phrine

-0.69

byss

-0.69

pressure

-0.68

ey

-0.66

irie

-0.64

undai

-0.64

anwhile

-0.63

POSITIVE LOGITS

asted

0.93

asts

0.85

 Minutes

0.81

ature

0.79

0.77

ovember

0.76

%-

0.75

OTAL

0.74

dfx

0.74

entric

0.73

Activations Density 0.032%

numbers representing ratings or votes

repeated numeric values, specifically voting tallies

No Comments

No Known Activations