INDEX

Explanations

titles or references to individuals with titles such as "Dr." and "Professor"

oai_token-act-pair · gpt-3.5-turbo

academic titles and their corresponding designations

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

 Stub

-0.80

ħĭ

-0.75

 Cambod

-0.71

uther

-0.70

 Greenwood

-0.69

susp

-0.65

Roy

-0.64

 cellul

-0.63

hiro

-0.62

tle

-0.61

POSITIVE LOGITS

2.62

2.44

2.17

1.88

Zo

1.62

Zer

1.59

Zig

1.57

Zar

1.57

zag

1.55

Zap

1.53

Activations Density 0.370%

titles or references to individuals with titles such as "Dr." and "Professor"

academic titles and their corresponding designations

No Comments

No Known Activations