INDEX

Explanations

mentions of groups of people or individuals in specific scenarios

oai_token-act-pair · gpt-3.5-turbo

references to unspecified groups or individuals in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

emonium

-0.78

ves

-0.68

 Sunshine

-0.64

 syndrome

-0.64

hap

-0.61

orde

-0.61

-0.60

 Isle

-0.59

 Syndrome

-0.58

 Debate

-0.57

POSITIVE LOGITS

 totaling

1.00

 notably

0.96

 originating

0.93

 representing

0.86

 prominently

0.86

 specializing

0.85

 comprising

0.85

 consisting

0.85

 citing

0.82

 weighing

0.82

Activations Density 0.125%

mentions of groups of people or individuals in specific scenarios

references to unspecified groups or individuals in various contexts

No Comments

No Known Activations