INDEX

Explanations

locations or proper nouns associated with specific places

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of places and organizations

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

 satisfaction

-0.64

 unfavorable

-0.60

 disposable

-0.59

 behav

-0.58

 craving

-0.58

 prejudice

-0.57

 favourable

-0.57

 suspic

-0.57

 overwhelming

-0.57

 favorable

-0.56

POSITIVE LOGITS

ornings

0.73

lore

0.68

itars

0.66

anky

0.64

esan

0.63

agent

0.63

jon

0.63

itus

0.61

wyn

0.60

selage

0.59

Activations Density 0.859%

locations or proper nouns associated with specific places

proper nouns, particularly names of places and organizations

No Comments

No Known Activations