INDEX

Explanations

phrases related to taking actions or making decisions

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate direction or inclination towards a specific outcome

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Awakens

-0.65

 rather

-0.65

 nutshell

-0.58

 Rather

-0.57

Ã¸

-0.56

pora

-0.56

 Offline

-0.55

rame

-0.55

apest

-0.55

 Meridian

-0.55

POSITIVE LOGITS

 sidx

0.66

 imaginable

0.65

Attach

0.64

anism

0.62

antage

0.61

ilers

0.60

depending

0.59

 concede

0.58

WARD

0.56

 approvals

0.56

Activations Density 1.016%

phrases related to taking actions or making decisions

phrases that indicate direction or inclination towards a specific outcome

No Comments

No Known Activations