INDEX

Explanations

the action of pointing something out

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate someone is making a statement or observation

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

asus

-0.82

wake

-0.80

ascus

-0.66

 experien

-0.65

oil

-0.64

avorite

-0.64

inally

-0.64

aden

-0.64

blood

-0.64

isol

-0.64

POSITIVE LOGITS

 loopholes

0.76

wards

0.72

 deductions

0.70

orial

0.68

rers

0.67

pad

0.67

how

0.66

IFF

0.65

DERR

0.65

:[

0.64

Activations Density 0.023%

the action of pointing something out

phrases that indicate someone is making a statement or observation

No Comments

No Known Activations