INDEX

Explanations

references to the name "Martin"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Martin" across various contexts

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

ngth

-0.91

ramid

-0.81

grades

-0.80

sie

-0.78

GoldMagikarp

-0.77

jriwal

-0.75

Ĥª

-0.73

ĳå£«

-0.73

riter

-0.70

elsius

-0.70

POSITIVE LOGITS

 Luther

1.49

ique

0.98

 Martin

0.88

 Sheen

0.88

ucci

0.83

etti

0.83

endez

0.82

ias

0.79

otti

0.75

 Hein

0.72

Activations Density 0.015%

references to the name "Martin"

mentions of the name "Martin" across various contexts

No Comments

No Known Activations