INDEX

Explanations

phrases involving titles or positions at companies

oai_token-act-pair · gpt-3.5-turbo

references to professional titles or roles in organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

SPONSORED

-0.89

Synopsis

-0.78

 skirts

-0.75

 copied

-0.69

 dressing

-0.68

perature

-0.66

 versa

-0.65

 resembled

-0.65

 somehow

-0.65

chwitz

-0.65

POSITIVE LOGITS

 consultancy

1.04

 Compass

0.96

 Consumers

0.93

 consulting

0.91

 Ceres

0.90

 Applied

0.87

 Concern

0.86

 Catalyst

0.84

 nonprofit

0.84

 Consulting

0.84

Activations Density 0.098%

phrases involving titles or positions at companies

references to professional titles or roles in organizations

No Comments

No Known Activations