INDEX

Explanations

mentions of and references to the location "Washington."

oai_token-act-pair · gpt-3.5-turbo

the word "wa" 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 @ 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

sis

-0.87

bol

-0.65

regor

-0.65

 Cell

-0.64

cu

-0.62

rations

-0.62

gon

-0.58

trip

-0.58

raph

-0.57

chini

-0.57

POSITIVE LOGITS

velength

1.70

veland

1.19

iting

1.18

iving

1.11

iver

1.02

isted

1.00

ivable

0.99

ived

0.99

ited

0.99

heed

0.99

Activations Density 0.044%

mentions of and references to the location "Washington."

the word "wa" in various contexts

No Comments

No Known Activations