INDEX

Explanations

proper nouns starting with the letter 'T'

oai_token-act-pair · gpt-3.5-turbo

acronyms and abbreviations related to organizations or categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 quar

-0.73

 sanct

-0.60

Nur

-0.58

 STATS

-0.57

ket

-0.54

 virtues

-0.54

 spat

-0.54

-0.53

 Mecca

-0.53

imov

-0.53

POSITIVE LOGITS

edo

1.07

oshenko

1.05

ulhu

0.92

flix

0.85

uyomi

0.81

tsky

0.79

hiba

0.78

dfx

0.78

asury

0.77

antage

0.74

Activations Density 0.136%

proper nouns starting with the letter 'T'

acronyms and abbreviations related to organizations or categories

No Comments

No Known Activations