INDEX

Explanations

links or references to additional information or content

oai_token-act-pair · gpt-3.5-turbo

instances of the word "can."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

burgh

-0.66

pires

-0.66

Moz

-0.66

 Yards

-0.61

phas

-0.61

 Mant

-0.61

Bridge

-0.59

 Appears

-0.59

 Passage

-0.59

 revisions

-0.59

POSITIVE LOGITS

't

1.49

NOT

1.14

 optionally

0.96

berra

0.92

 easily

0.91

 choose

0.90

isters

0.90

ister

0.88

 customize

0.87

 activate

0.87

Activations Density 0.093%

links or references to additional information or content

instances of the word "can."

No Comments

No Known Activations