INDEX

Explanations

phrases indicating or suggesting a specific conclusion or outcome

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

enium

-0.80

agues

-0.77

quer

-0.75

ighth

-0.73

-0.69

pez

-0.68

az

-0.66

uristic

-0.65

ãĤ´ãĥ³

-0.65

anny

-0.65

POSITIVE LOGITS

 displeasure

0.94

 otherwise

0.93

 that

0.87

 dissatisfaction

0.86

 competence

0.85

 familiarity

0.85

 impending

0.84

 imminent

0.83

 willingness

0.83

 impat

0.82

Activations Density 0.112%

phrases indicating or suggesting a specific conclusion or outcome

No Comments

No Known Activations