INDEX

Explanations

references to specific organizations, individuals, and dates

oai_token-act-pair · gpt-3.5-turbo

specific names or titles related to organizations and positions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 tricks

-0.85

 wouldn

-0.84

 chatter

-0.82

toggle

-0.81

 trickle

-0.81

lur

-0.79

 slam

-0.79

doesn

-0.76

 tweaks

-0.76

 pricey

-0.76

POSITIVE LOGITS

 pursuant

1.14

.''.

1.09

.[

1.03

([

1.02

.(

1.01

 Fiscal

0.90

''.

0.90

!.

0.90

 Limited

0.89

 whilst

0.89

Activations Density 0.569%

references to specific organizations, individuals, and dates

specific names or titles related to organizations and positions

No Comments

No Known Activations