INDEX

Explanations

mentions of fictional detective characters, particularly Sherlock Holmes

oai_token-act-pair · gpt-3.5-turbo

mentions of the character Sherlock Holmes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ital

-0.78

egal

-0.74

ribute

-0.70

AFTA

-0.69

abet

-0.68

ishers

-0.68

leased

-0.67

kick

-0.67

endants

-0.66

inished

-0.65

POSITIVE LOGITS

 Holmes

1.20

Lens

0.84

iday

0.78

combe

0.75

 Manor

0.71

 Sherlock

0.70

 Eleven

0.70

 Reloaded

0.69

 TAMADRA

0.69

 Circus

0.68

Activations Density 0.010%

mentions of fictional detective characters, particularly Sherlock Holmes

mentions of the character Sherlock Holmes

No Comments

No Known Activations