INDEX

Explanations

proper nouns, specifically names like "Lawrence."

oai_token-act-pair · gpt-3.5-turbo

the repeated mention of the name "Lawrence" 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 @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

hedral

-1.00

graded

-0.89

ramid

-0.87

inately

-0.77

scribe

-0.76

minist

-0.75

uttered

-0.73

iliar

-0.73

arding

-0.71

cffff

-0.70

POSITIVE LOGITS

 Liver

1.12

 Hague

0.84

 Berkeley

0.83

Wel

0.82

Kra

0.82

ville

0.81

rence

0.81

burg

0.79

 Summers

0.77

 Lawrence

0.76

Activations Density 0.041%

proper nouns, specifically names like "Lawrence."

the repeated mention of the name "Lawrence" in various contexts

No Comments

No Known Activations