INDEX

Explanations

phrases indicating minimum requirements or thresholds

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

statements indicating a minimum requirement or threshold

oai_token-act-pair · gpt-3.5-turbo

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tions

-0.75

Reviewer

-0.70

quit

-0.67

oris

-0.66

 Dynamics

-0.63

rats

-0.62

ãĤ¿

-0.61

ues

-0.60

icides

-0.60

Generic

-0.59

POSITIVE LOGITS

 partially

0.88

 partly

0.82

 moderately

0.74

uner

0.74

lik

0.73

 SOME

0.71

 toler

0.70

 temporarily

0.70

 mildly

0.68

 marginally

0.67

Activations Density 0.033%

phrases indicating minimum requirements or thresholds

statements indicating a minimum requirement or threshold

No Comments

No Known Activations