INDEX

Explanations

locations/geographical terms

oai_token-act-pair · gpt-3.5-turbo

geographical locations and entities related to specific regions

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

 Canaver

-0.68

ï¸ı

-0.63

ongyang

-0.58

-+-+

-0.57

unctions

-0.56

together

-0.55

lessness

-0.54

vind

-0.54

 Haram

-0.53

utterstock

-0.53

POSITIVE LOGITS

 equivalents

0.65

 Blvd

0.58

 equivalent

0.57

 Express

0.55

Rd

0.55

 Classics

0.54

 edition

0.51

 stereo

0.51

 Theme

0.50

 Style

0.50

Activations Density 0.567%

locations/geographical terms

geographical locations and entities related to specific regions

No Comments

No Known Activations