INDEX

Explanations

proper nouns in various contexts

oai_token-act-pair · gpt-3.5-turbo

proper nouns, specifically names of individuals or entities

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

 dstg

-0.74

 guiName

-0.66

CCC

-0.60

Higher

-0.59

Effective

-0.59

manufact

-0.59

taboola

-0.58

 EDITION

-0.58

vironment

-0.56

POSITIVE LOGITS

 recalled

0.83

 recalls

0.83

 remembers

0.79

 recounts

0.78

 recounted

0.78

 wrote

0.75

 laughed

0.73

 confessed

0.73

 admitted

0.72

 admits

0.71

Activations Density 0.407%

proper nouns in various contexts

proper nouns, specifically names of individuals or entities

No Comments

No Known Activations