INDEX

Explanations

phrases related to being at risk

oai_token-act-pair · gpt-3.5-turbo

phrases indicating potential hazards or dangers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

anche

-0.81

rix

-0.76

elf

-0.67

ript

-0.65

ovy

-0.65

 headquartered

-0.64

ao

-0.63

ann

-0.62

erent

-0.62

MX

-0.61

POSITIVE LOGITS

ħĭ

0.84

 endanger

0.84

angering

0.75

 undermin

0.74

 casualties

0.72

 schizophren

0.70

 Survive

0.68

 dism

0.68

 pitfalls

0.68

 starvation

0.67

Activations Density 0.022%

phrases related to being at risk

phrases indicating potential hazards or dangers

No Comments

No Known Activations