INDEX

Explanations

phrases indicating permission or allowance for certain actions or situations

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate permission or the ability to do something

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

 Originally

-0.72

soever

-0.70

 Notting

-0.70

edin

-0.69

 Dreams

-0.67

 Carlton

-0.65

DOWN

-0.63

ynski

-0.62

lime

-0.62

 Slug

-0.61

POSITIVE LOGITS

gery

0.84

geries

0.78

 reinforcements

0.77

bidden

0.74

 extra

0.74

ocal

0.74

rocal

0.73

 speedy

0.73

 greater

0.72

 additional

0.71

Activations Density 0.065%

phrases indicating permission or allowance for certain actions or situations

phrases that indicate permission or the ability to do something

No Comments

No Known Activations