INDEX

Explanations

terms related to legal and judicial processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res_post_32k-oai

Configuration

jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

32,768

Data Type

torch.float32

Hook Name

blocks.10.hook_resid_post

Hook Layer

Architecture

standard

Context Size

Dataset

Skylion007/openwebtext

Activation Function

topk

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

LV

-0.80

 Nebula

-0.77

anchester

-0.72

 Hazel

-0.72

Osw

-0.71

 Sutherland

-0.69

 Loch

-0.67

apore

-0.67

 Lauder

-0.67

Yel

-0.67

POSITIVE LOGITS

 facto

0.95

 precedent

0.79

 excuse

0.74

 oper

0.72

 error

0.71

 lett

0.70

 comprom

0.69

 contro

0.68

 argument

0.68

endi

0.65

Activations Density 0.498%

terms related to legal and judicial processes

No Comments

No Known Activations