INDEX

Explanations

reasons or justifications for certain actions or phenomena

oai_token-act-pair · gpt-3.5-turbo

questions and justifications for reasons

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

ivalent

-0.75

isha

-0.73

ona

-0.73

urated

-0.72

iership

-0.72

ail

-0.71

lude

-0.68

 favourites

-0.67

 nails

-0.67

nik

-0.66

POSITIVE LOGITS

 scarcity

0.79

 misunderstand

0.78

ecause

0.77

 Because

0.76

 Firstly

0.76

arcity

0.73

Because

0.71

 incent

0.68

cause

0.67

 antitrust

0.65

Activations Density 0.270%

reasons or justifications for certain actions or phenomena

questions and justifications for reasons

No Comments

No Known Activations