INDEX

Explanations

words related to policies, regulations, and political actions

oai_token-act-pair · gpt-3.5-turbo

statements that reference the effects and roles of something significant, often using the word "it."

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

ombo

-0.72

IFE

-0.72

english

-0.66

onica

-0.66

uga

-0.65

itu

-0.65

agon

-0.64

isphere

-0.63

 Albion

-0.62

ibble

-0.62

POSITIVE LOGITS

 also

0.97

 encourage

0.91

 encourages

0.89

 allow

0.86

 additionally

0.86

 eliminate

0.85

 instruct

0.83

 facilitate

0.83

 furthermore

0.83

 thereby

0.82

Activations Density 0.489%

words related to policies, regulations, and political actions

statements that reference the effects and roles of something significant, often using the word "it."

No Comments

No Known Activations