INDEX

Explanations

specific words or phrases related to legal discussions or procedures

oai_token-act-pair · gpt-3.5-turbo

affirmations and positive responses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ectomy

-0.58

clave

-0.57

 Kore

-0.57

 Marshall

-0.56

Mun

-0.56

weights

-0.56

 Regulation

-0.55

 Brooks

-0.54

 Miliband

-0.53

 Kodi

-0.53

POSITIVE LOGITS

anwhile

0.85

etheless

0.77

 dunno

0.73

interrupted

0.69

[/

0.68

onis

0.68

xon

0.67

imately

0.66

seless

0.66

Ü

0.66

Activations Density 0.175%

specific words or phrases related to legal discussions or procedures

affirmations and positive responses

No Comments

No Known Activations