INDEX

Explanations

places, locations, or institutions within a sentence or document

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

proper nouns, specifically names of people and places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

âĶĢâĶĢâĶĢâĶĢ

-0.64

 cull

-0.56

ARY

-0.56

aries

-0.55

 nothing

-0.55

 test

-0.55

arist

-0.55

Playing

-0.54

chell

-0.53

 emulate

-0.53

POSITIVE LOGITS

imb

0.85

enes

0.84

ene

0.83

ois

0.80

apon

0.80

ennes

0.79

obos

0.79

idal

0.79

ogh

0.76

enf

0.75

Activations Density 0.279%

places, locations, or institutions within a sentence or document

proper nouns, specifically names of people and places

No Comments

No Known Activations