INDEX

Explanations

variations or differences in quantities, proportions, or characteristics in various contexts

oai_token-act-pair · gpt-3.5-turbo

terms related to variability and differences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

odore

-0.81

tur

-0.78

zyk

-0.73

stra

-0.71

swick

-0.71

Bern

-0.68

tri

-0.68

gets

-0.68

TE

-0.67

oops

-0.65

POSITIVE LOGITS

 wildly

1.10

 degrees

1.10

 depending

1.02

 widely

0.99

 markedly

0.98

 geographically

0.97

 between

0.96

 considerably

0.95

 greatly

0.90

 enormously

0.90

Activations Density 0.064%

variations or differences in quantities, proportions, or characteristics in various contexts

terms related to variability and differences

No Comments

No Known Activations