INDEX

Explanations

verbs related to demonstrating or proving something, often with high confidence

oai_token-act-pair · gpt-3.5-turbo

instances of the word "shown" or its variations indicating evidence or demonstration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

etheless

-0.75

agher

-0.68

assisted

-0.66

nian

-0.63

srf

-0.63

osate

-0.62

avascript

-0.62

ading

-0.62

ascus

-0.62

alez

-0.61

POSITIVE LOGITS

ered

1.03

manship

0.93

runners

0.80

 signs

0.78

alter

0.76

biz

0.74

\\\\\\\\

0.71

 flashes

0.70

eele

0.69

heet

0.68

Activations Density 0.060%

verbs related to demonstrating or proving something, often with high confidence

instances of the word "shown" or its variations indicating evidence or demonstration

No Comments

No Known Activations