INDEX

Explanations

names related to people or places

oai_token-act-pair · gpt-3.5-turbo

proper nouns and names associated with people or places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

å°Ĩ

-0.76

PsyNetMessage

-0.68

andro

-0.64

)=(

-0.61

ãĥ¯

-0.61

osph

-0.60

æ©

-0.60

ãĥ¼ãĥĨãĤ£

-0.59

 PACs

-0.57

VID

-0.57

POSITIVE LOGITS

istan

0.87

jri

0.83

wrap

0.77

orthy

0.77

culosis

0.76

ashi

0.75

lease

0.73

zee

0.71

anski

0.70

ji

0.69

Activations Density 0.308%

names related to people or places

proper nouns and names associated with people or places

No Comments

No Known Activations