INDEX

Explanations

phrases related to conjuring or imagining something

oai_token-act-pair · gpt-3.5-turbo

terms related to creating or invoking ideas and concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

imura

-0.80

ocene

-0.70

iary

-0.70

amoto

-0.68

ow

-0.66

hill

-0.66

esta

-0.65

 Rebellion

-0.64

 variance

-0.62

 exception

-0.61

POSITIVE LOGITS

urer

0.88

ure

0.83

unct

0.81

urers

0.77

eger

0.76

uration

0.76

uel

0.73

ãĥ¥

0.73

quel

0.72

uss

0.70

Activations Density 0.067%

phrases related to conjuring or imagining something

terms related to creating or invoking ideas and concepts

No Comments

No Known Activations