INDEX

Explanations

specific roles or titles related to professional positions

oai_token-act-pair · gpt-3.5-turbo

references to positions or roles related to directorial responsibilities

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Egyptians

-0.79

 Greeks

-0.74

theless

-0.70

Joined

-0.69

Tokens

-0.65

é¾įå

-0.63

onyms

-0.62

Ü

-0.62

 horr

-0.62

 tumble

-0.60

POSITIVE LOGITS

 communications

0.97

orney

0.88

forestation

0.86

aganda

0.85

ongyang

0.83

 communication

0.79

research

0.77

 Strategic

0.76

roit

0.76

 research

0.75

Activations Density 0.051%

specific roles or titles related to professional positions

references to positions or roles related to directorial responsibilities

No Comments

No Known Activations