INDEX

Explanations

references to adaptations of stories or content into other media forms

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

ciating

-0.79

 diagram

-0.71

dash

-0.67

aurus

-0.66

 waivers

-0.65

 diagrams

-0.64

 Github

-0.63

 Newsletter

-0.63

 disposal

-0.62

pite

-0.62

POSITIVE LOGITS

ãĤ´

0.78

 starring

0.70

 Artemis

0.68

 Milton

0.67

 Franch

0.67

 Friedrich

0.66

enegger

0.65

 lett

0.64

ownt

0.64

 corrid

0.64

Activations Density 0.862%

references to adaptations of stories or content into other media forms

No Comments

No Known Activations