INDEX

Explanations

targets or entities that are being identified or aimed at

oai_token-act-pair · gpt-3.5-turbo

references to specific objectives or targets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

OVA

-0.71

haz

-0.68

pole

-0.66

cup

-0.65

john

-0.65

htt

-0.65

alus

-0.63

uitive

-0.62

inct

-0.62

inus

-0.62

POSITIVE LOGITS

 targets

4.09

 target

2.72

target

2.09

Target

1.83

 targeted

1.82

 Targ

1.79

 targeting

1.74

 Target

1.70

 targ

1.66

 objectives

1.61

Activations Density 0.009%

targets or entities that are being identified or aimed at

references to specific objectives or targets

No Comments

No Known Activations