INDEX

Explanations

questions and answers (Q&A) related text

oai_token-act-pair · gpt-3.5-turbo

mentions of Q&A sessions or discussions

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

ufact

-0.83

combe

-0.77

adr

-0.73

wcs

-0.72

jiang

-0.70

pend

-0.70

rug

-0.69

 retina

-0.69

cms

-0.67

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.66

POSITIVE LOGITS

 session

0.77

Cooldown

0.69

 narration

0.69

 Commentary

0.68

 consultations

0.68

 booth

0.67

 interview

0.67

FAQ

0.65

 sessions

0.65

 guru

0.65

Activations Density 0.029%

questions and answers (Q&A) related text

mentions of Q&A sessions or discussions

No Comments

No Known Activations