INDEX

Explanations

references to additional items or concepts typically introduced by "among other things."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res_post_32k-oai

Configuration

jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_6.pt

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

32,768

Data Type

torch.float32

Hook Name

blocks.6.hook_resid_post

Hook Layer

Architecture

standard

Context Size

Dataset

Skylion007/openwebtext

Activation Function

topk

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ocene

-0.75

ature

-0.68

 Asheville

-0.67

 civilisation

-0.66

ocent

-0.65

 Anfield

-0.64

OK

-0.64

ok

-0.63

DK

-0.61

Nun

-0.60

POSITIVE LOGITS

LEASE

0.80

 includ

0.78

features

0.76

interstitial

0.75

Øª

0.72

accompan

0.69

hips

0.65

Ø©

0.65

properties

0.65

Ø³

0.65

Activations Density 0.083%

references to additional items or concepts typically introduced by "among other things."

No Comments

No Known Activations