INDEX

Explanations

words related to causing a reaction or inciting action

oai_token-act-pair · gpt-3.5-turbo

language associated with inciting or provoking actions and emotions

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

 Equipment

-0.64

 rendered

-0.64

é¾įå

-0.63

 Drawn

-0.62

BILITY

-0.61

 Applic

-0.59

WARE

-0.59

 ported

-0.58

 inspected

-0.58

breaker

-0.58

POSITIVE LOGITS

 passions

1.18

 controversy

1.18

 resentment

1.12

 curiosity

1.09

 hatred

1.08

 enthusiasm

1.07

 excitement

1.07

 suspicions

1.06

 discontent

1.06

 outrage

1.05

Activations Density 0.237%

words related to causing a reaction or inciting action

language associated with inciting or provoking actions and emotions

No Comments

No Known Activations