INDEX

Explanations

instances of the word "letters" in texts

oai_token-act-pair · gpt-3.5-turbo

references to letters 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 @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

amera

-0.76

rolet

-0.74

 WATCH

-0.69

Latest

-0.69

ESCO

-0.66

oleon

-0.66

lace

-0.64

Govern

-0.63

invoke

-0.63

âķĲâķĲ

-0.63

POSITIVE LOGITS

 letters

1.20

mith

1.10

letter

0.99

 letter

0.90

acters

0.89

Letter

0.87

worms

0.84

ername

0.84

 lett

0.81

ymes

0.80

Activations Density 0.014%

instances of the word "letters" in texts

references to letters in various contexts

No Comments

No Known Activations