INDEX

Explanations

words related to scandals, controversies, and negative events involving individuals or organizations

oai_token-act-pair · gpt-3.5-turbo

negative descriptors for people or groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 process

-0.79

arching

-0.77

icum

-0.77

phasis

-0.76

 holster

-0.72

 contrace

-0.72

 footprint

-0.72

idation

-0.71

 tremend

-0.71

 incent

-0.70

POSITIVE LOGITS

Lt

0.93

 newcomer

0.92

 extraord

0.92

 businessman

0.90

Sgt

0.87

 Tony

0.86

 billionaire

0.85

Sen

0.85

 celebrities

0.85

 journalist

0.85

Activations Density 0.560%

words related to scandals, controversies, and negative events involving individuals or organizations

negative descriptors for people or groups

No Comments

No Known Activations