INDEX

Explanations

phrases indicating action or occurrence

oai_token-act-pair · gpt-3.5-turbo

expressions of concern and observations about situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SHIP

-0.74

 Horses

-0.71

etheless

-0.69

 Cards

-0.65

 âĹı

-0.65

existence

-0.65

istries

-0.65

 Drive

-0.64

 Grape

-0.63

 Carnival

-0.63

POSITIVE LOGITS

 progresses

0.84

 progressed

0.81

 unfolds

0.70

 prepares

0.68

 matured

0.67

 instructor

0.66

 wont

0.65

 unfolded

0.65

 approached

0.64

 traditionally

0.64

Activations Density 0.588%

phrases indicating action or occurrence

expressions of concern and observations about situations

No Comments

No Known Activations