INDEX

Explanations

mentions of the name "Alice"

oai_token-act-pair · gpt-3.5-turbo

references to the character Alice, particularly in relation to "Alice in Wonderland."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

enegger

-0.93

ODUCT

-0.80

ework

-0.76

soDeliveryDate

-0.75

unct

-0.67

racted

-0.66

eur

-0.65

BOOK

-0.65

ulates

-0.64

iaries

-0.63

POSITIVE LOGITS

Alice

1.00

 Alice

0.95

 Springs

0.89

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.79

pheus

0.79

uria

0.78

 Weir

0.77

 Guth

0.76

 Wonderland

0.76

isle

0.75

Activations Density 0.005%

mentions of the name "Alice"

references to the character Alice, particularly in relation to "Alice in Wonderland."

No Comments

No Known Activations