INDEX

Explanations

mentions of the word "savage"

oai_token-act-pair · gpt-3.5-turbo

references to the term "Savage" and its variations, often in the context of storytelling or descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

orescent

-1.10

ership

-0.91

unal

-0.86

psey

-0.86

olulu

-0.85

ipt

-0.80

uting

-0.79

oard

-0.78

osph

-0.76

graded

-0.75

POSITIVE LOGITS

 Savage

1.02

 Carlson

0.75

 DRAG

0.75

 Royale

0.73

 Unleashed

0.72

 asses

0.70

 Rack

0.69

 Hogan

0.69

 Scotia

0.69

inelli

0.67

Activations Density 0.009%

mentions of the word "savage"

references to the term "Savage" and its variations, often in the context of storytelling or descriptions

No Comments

No Known Activations