INDEX

Explanations

words related to safety and security

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing the concept of safety

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

amy

-0.73

iery

-0.70

 betrayal

-0.69

 willingness

-0.65

 newsletters

-0.64

yi

-0.63

enf

-0.63

ilion

-0.63

 directions

-0.62

 Killer

-0.61

POSITIVE LOGITS

 conclud

0.88

 exting

0.80

 dispose

0.80

mint

0.79

 transitioned

0.77

ufact

0.76

 evacuated

0.72

ãĤ©

0.72

 aver

0.70

 evacuate

0.69

Activations Density 0.026%

words related to safety and security

phrases emphasizing the concept of safety

No Comments

No Known Activations