INDEX

Explanations

instances of comparisons suggesting a decrease or reduction in magnitude

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a comparison or measure of quantity

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

TRY

-0.69

 Reconstruction

-0.66

 Origins

-0.66

DD

-0.61

AE

-0.61

den

-0.61

âĹ¼

-0.60

kamp

-0.57

DK

-0.57

RL

-0.57

POSITIVE LOGITS

ened

0.98

 than

0.96

ening

0.87

thumbnails

0.82

 Than

0.75

ainers

0.72

expensive

0.72

ons

0.70

ensive

0.69

 fortunate

0.69

Activations Density 0.032%

instances of comparisons suggesting a decrease or reduction in magnitude

phrases indicating a comparison or measure of quantity

No Comments

No Known Activations