INDEX

Explanations

phrases indicating consequences or implications

oai_token-act-pair · gpt-3.5-turbo

expressions that indicate a result or consequence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

soType

-0.73

 Examples

-0.68

Suggest

-0.67

mbuds

-0.67

ugen

-0.63

Winged

-0.62

Newsletter

-0.62

ealous

-0.62

 Saud

-0.60

ruff

-0.59

POSITIVE LOGITS

 potentially

0.85

 effectively

0.84

 fewer

0.84

 sooner

0.79

 automatically

0.76

 avoided

0.76

 unavoid

0.75

 liable

0.75

pora

0.75

 likely

0.75

Activations Density 0.277%

phrases indicating consequences or implications

expressions that indicate a result or consequence

No Comments

No Known Activations