INDEX

Explanations

words related to a specific location or organization, specifically "BC"

oai_token-act-pair · gpt-3.5-turbo

references to "BC" as it relates to various contexts, particularly historical and governmental

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

mble

-1.15

ãĤ¯

-0.89

lihood

-0.85

ktop

-0.82

strument

-0.77

ãĥŀ

-0.77

ãĥł

-0.74

éĹĺ

-0.74

ãĥĻ

-0.74

Dir

-0.73

POSITIVE LOGITS

BC

0.81

 Hydro

0.78

AMP

0.78

NDP

0.76

orne

0.76

RA

0.76

RC

0.75

BG

0.75

rypt

0.74

ategory

0.73

Activations Density 0.006%

words related to a specific location or organization, specifically "BC"

references to "BC" as it relates to various contexts, particularly historical and governmental

No Comments

No Known Activations