INDEX

Explanations

phrases introduced by the phrase "calls"

oai_token-act-pair · gpt-3.5-turbo

references to quotes or terms attributed to individuals

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ratulations

-0.58

umber

-0.57

haven

-0.57

eman

-0.56

agar

-0.56

iership

-0.55

ourney

-0.55

application

-0.54

ondo

-0.54

arnaev

-0.54

POSITIVE LOGITS

1.00

"'

0.95

''

0.87

0.83

 "âĢ¦

0.82

"#

0.81

 "...

0.80

"[

0.79

``

0.78

"_

0.76

Activations Density 0.088%

phrases introduced by the phrase "calls"

references to quotes or terms attributed to individuals

No Comments

No Known Activations