INDEX

Explanations

mentions of numerical values associated with physical characteristics such as density and thresholds

oai_token-act-pair · gpt-3.5-turbo

references to density in various contexts

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

akia

-0.85

bara

-0.83

uberty

-0.81

pha

-0.77

porary

-0.75

 McDonnell

-0.72

utic

-0.69

udos

-0.69

 WATCHED

-0.68

cffffcc

-0.67

POSITIVE LOGITS

 density

0.92

density

0.76

icity

0.73

 clust

0.72

 eater

0.71

ikuman

0.71

 census

0.70

 foam

0.70

ensity

0.69

layer

0.69

Activations Density 0.022%

mentions of numerical values associated with physical characteristics such as density and thresholds

references to density in various contexts

No Comments

No Known Activations