INDEX

Explanations

phrases indicating approval or acceptance

oai_token-act-pair · gpt-3.5-turbo

instances of the symbol 'âĢ'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 reception

-0.71

 elig

-0.69

Syd

-0.66

 sidx

-0.65

 shroud

-0.63

 carriage

-0.62

 Somerset

-0.62

omorphic

-0.62

 scattering

-0.61

 transitional

-0.61

POSITIVE LOGITS

¬

1.22

Ļ

1.00

¹

0.96

Ĵ

0.95

elong

0.94

º

0.91

¡

0.91

Ķ

0.90

¥

0.90

¤

0.89

Activations Density 0.333%

phrases indicating approval or acceptance

instances of the symbol 'âĢ'

No Comments

No Known Activations