INDEX

Explanations

mentions of significant changes or transformations

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "shift" indicating changes or transformations

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

amina

-0.67

gdala

-0.66

ecause

-0.66

ournals

-0.66

gom

-0.65

CRIP

-0.63

 Smile

-0.62

jug

-0.62

Interstitial

-0.61

ORED

-0.60

POSITIVE LOGITS

 gears

1.24

 toward

1.02

 towards

1.00

 shift

0.94

 away

0.94

 sands

0.93

 blame

0.89

 downwards

0.85

 shifts

0.82

shift

0.78

Activations Density 0.041%

mentions of significant changes or transformations

occurrences of the word "shift" indicating changes or transformations

No Comments

No Known Activations