INDEX

Explanations

proper nouns related to locations, specifically parks and geographical names

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

ching

-0.83

hiba

-0.72

oldown

-0.72

thia

-0.70

sembly

-0.67

gettable

-0.66

vati

-0.65

 algae

-0.65

 forwarding

-0.65

hement

-0.63

POSITIVE LOGITS

ÃĽ

0.71

 GOODMAN

0.70

dash

0.69

 Creed

0.68

 Stories

0.65

èĢħ

0.64

 Kimber

0.64

 Johannes

0.63

 Remastered

0.63

nation

0.62

Activations Density 0.065%

proper nouns related to locations, specifically parks and geographical names

No Comments

No Known Activations