INDEX

Explanations

phrases that introduce examples or samples

oai_token-act-pair · gpt-3.5-turbo

examples and instances used for clarification or illustration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

afort

-0.67

 loopholes

-0.66

 Ukrain

-0.66

 unaccount

-0.65

parency

-0.63

pmwiki

-0.62

negie

-0.62

utsu

-0.61

 unofficial

-0.61

ival

-0.61

POSITIVE LOGITS

foo

1.05

foo

1.05

XY

0.94

example

0.87

 Suppose

0.87

Foo

0.77

 hello

0.73

\(

0.72

 apple

0.69

 suppose

0.66

Activations Density 1.084%

phrases that introduce examples or samples

examples and instances used for clarification or illustration

No Comments

No Known Activations