INDEX

Explanations

instances of the phrase "to see what" followed by an action or result

oai_token-act-pair · gpt-3.5-turbo

questions and inquiries related to curiosity and exploration

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

UME

-0.73

pg

-0.70

cffffcc

-0.70

ufact

-0.67

stated

-0.67

Saharan

-0.64

 Caption

-0.64

agonists

-0.62

ometown

-0.62

 Verse

-0.62

POSITIVE LOGITS

 positives

0.64

 sparks

0.64

 whats

0.64

 mockery

0.63

erson

0.62

retty

0.62

 resil

0.61

 results

0.61

anian

0.61

 starve

0.59

Activations Density 0.091%

instances of the phrase "to see what" followed by an action or result

questions and inquiries related to curiosity and exploration

No Comments

No Known Activations