INDEX

Explanations

words related to challenges or disturbances in a system or process

oai_token-act-pair · gpt-3.5-turbo

references to disruption

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

-0.78

onz

-0.78

rouse

-0.76

ented

-0.73

estone

-0.70

Pres

-0.69

vet

-0.68

 Helpful

-0.68

cop

-0.67

gets

-0.66

POSITIVE LOGITS

 disruption

1.18

 disruptions

1.09

 interruption

1.04

 outage

0.87

 disrupted

0.83

 havoc

0.83

 inconvenience

0.81

xual

0.80

 inconven

0.79

cov

0.74

Activations Density 0.015%

words related to challenges or disturbances in a system or process

references to disruption

No Comments

No Known Activations