INDEX

Explanations

adjectives describing steepness or difficulty

oai_token-act-pair · gpt-3.5-turbo

descriptors of steepness or difficulty

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ignty

-0.79

uve

-0.71

 IMAGES

-0.71

uter

-0.70

umbn

-0.69

gdala

-0.69

ubb

-0.67

 Holo

-0.65

Smy

-0.62

bers

-0.62

POSITIVE LOGITS

 slopes

1.12

ly

1.08

 slope

1.08

ened

1.05

ening

1.03

 climb

0.96

 gradient

0.93

 uphill

0.91

 ascent

0.87

nesses

0.86

Activations Density 0.054%

adjectives describing steepness or difficulty

descriptors of steepness or difficulty

No Comments

No Known Activations