INDEX

Explanations

instances of transitions between different phrases or concepts

oai_token-act-pair · gpt-3.5-turbo

phrases that introduce paraphrased statements or explanations

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

oly

-0.79

ometer

-0.71

igators

-0.64

roy

-0.64

 crocod

-0.62

aciously

-0.61

arov

-0.61

minster

-0.60

ometers

-0.59

atars

-0.59

POSITIVE LOGITS

 ACTIONS

0.73

"[

0.71

 unless

0.70

FTWARE

0.68

Interested

0.68

 barring

0.68

 excluding

0.66

yes

0.65

 Ø§ÙĦ

0.64

 yeah

0.64

Activations Density 0.145%

instances of transitions between different phrases or concepts

phrases that introduce paraphrased statements or explanations

No Comments

No Known Activations