INDEX

Explanations

specific sequences of terms and characters

oai_token-act-pair · gpt-3.5-turbo

negations, particularly focusing on the phrase "not a."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

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

outhern

-0.65

usions

-0.61

////////////////

-0.61

razil

-0.59

ugu

-0.59

arer

-0.57

////////////////////////////////

-0.55

avez

-0.55

iliary

-0.55

atu

-0.55

POSITIVE LOGITS

 rake

0.53

 Pony

0.48

 Crunch

0.47

="#

0.47

 kitten

0.46

CPC

0.45

 Slime

0.45

 pony

0.45

(/

0.44

tee

0.44

Activations Density 1.101%

specific sequences of terms and characters

negations, particularly focusing on the phrase "not a."

No Comments

No Known Activations