INDEX

Explanations

the word "mean" or its variations in the context of implying a consequence or interpretation

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate causality or implications

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

@#&

-0.82

atech

-0.78

thumbnails

-0.78

oqu

-0.76

uded

-0.72

itations

-0.70

iazep

-0.70

phrine

-0.69

taboola

-0.68

iar

-0.68

POSITIVE LOGITS

 goodbye

1.12

 sacrificing

0.94

 nothing

0.91

 abandoning

0.86

 something

0.81

 risking

0.81

 fewer

0.78

 avoiding

0.77

 anything

0.74

 that

0.71

Activations Density 0.045%

the word "mean" or its variations in the context of implying a consequence or interpretation

phrases that indicate causality or implications

No Comments

No Known Activations