INDEX

Explanations

proper nouns or specific entities described by different individuals

oai_token-act-pair · gpt-3.5-turbo

attributions and references to various sources, such as witnesses, pundits, and authorities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

acca

-0.63

arching

-0.61

ï¸ı

-0.57

 unloaded

-0.57

orate

-0.57

 confronted

-0.56

 PLEASE

-0.55

OPA

-0.55

:]

-0.55

SU

-0.55

POSITIVE LOGITS

 alike

0.85

 circles

0.84

DERR

0.74

glers

0.72

pedia

0.70

genre

0.70

 acclaim

0.70

 initials

0.69

 aliases

0.67

ãĥı

0.66

Activations Density 0.307%

proper nouns or specific entities described by different individuals

attributions and references to various sources, such as witnesses, pundits, and authorities

No Comments

No Known Activations