INDEX

Explanations

quotes where someone is saying something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "said" and its variations in verbal discourse

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

phia

-0.78

berus

-0.66

 Explosion

-0.61

GV

-0.60

arent

-0.59

ppings

-0.59

 Potential

-0.58

actly

-0.57

azines

-0.57

Fit

-0.57

POSITIVE LOGITS

 confidently

0.99

 sarcast

0.98

 emphatically

0.98

 passionately

0.96

 proudly

0.95

 bluntly

0.92

 calmly

0.92

 enthusiastically

0.90

 bitterly

0.89

 angrily

0.88

Activations Density 0.097%

quotes where someone is saying something

instances of the word "said" and its variations in verbal discourse

No Comments

No Known Activations