INDEX

Explanations

references to related stories or articles

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

 surv

-0.65

atron

-0.61

ACY

-0.60

 gren

-0.59

LOD

-0.59

gran

-0.59

ASED

-0.59

wagen

-0.57

 photons

-0.57

 normally

-0.57

POSITIVE LOGITS

 Learns

0.77

 Expand

0.75

anguage

0.71

andise

0.70

reddits

0.69

aughs

0.67

 Cosponsors

0.65

ideos

0.65

pac

0.65

 Legislation

0.64

Activations Density 0.103%

related news article headlines

references to related stories or articles

No Comments

No Known Activations