INDEX

Explanations

questions starting with the word "Who"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Who" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

MER

-0.90

interstitial

-0.73

pit

-0.71

PORT

-0.71

 Rath

-0.69

 Hyde

-0.68

Roy

-0.67

RAL

-0.67

rations

-0.66

UTION

-0.65

POSITIVE LOGITS

soever

1.17

oping

0.86

 else

0.86

oped

0.83

ispers

0.81

 cares

0.81

resy

0.80

ever

0.79

 redes

0.78

 blinked

0.77

Activations Density 0.076%

questions starting with the word "Who"

instances of the word "Who" in various contexts

No Comments

No Known Activations