INDEX

Explanations

phrases related to the capabilities or potential of entities or products

oai_token-act-pair · gpt-3.5-turbo

phrases indicating capabilities and the impact of actions or events

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

estern

-0.68

onet

-0.61

Vil

-0.59

noticed

-0.58

 Hills

-0.56

 surn

-0.56

 Diesel

-0.56

 urinary

-0.56

uclear

-0.55

 briefs

-0.55

POSITIVE LOGITS

 entail

0.99

pires

0.84

 entails

0.81

ĸļ

0.69

SourceFile

0.68

.(

0.66

incial

0.65

ablishment

0.65

 purported

0.65

 consist

0.65

Activations Density 0.203%

phrases related to the capabilities or potential of entities or products

phrases indicating capabilities and the impact of actions or events

No Comments

No Known Activations