INDEX

Explanations

the name "Trent" with various surrounding context

oai_token-act-pair · gpt-3.5-turbo

mentions of the character "Trent."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

drm

-0.76

wl

-0.69

soType

-0.68

rm

-0.67

arettes

-0.66

overs

-0.65

draft

-0.64

arette

-0.62

java

-0.62

anticipated

-0.62

POSITIVE LOGITS

 Trent

1.25

Rez

1.24

orial

0.84

urous

0.77

anus

0.77

olini

0.73

 Baal

0.72

seys

0.72

 Richardson

0.72

Dil

0.71

Activations Density 0.009%

the name "Trent" with various surrounding context

mentions of the character "Trent."

No Comments

No Known Activations