INDEX

Explanations

phrases referring to directions and positions

oai_token-act-pair · gpt-3.5-turbo

terms related to flatness and dimensionality in objects or concepts

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

 SERV

-0.78

 Cosponsors

-0.65

è¦ļéĨĴ

-0.64

alty

-0.64

 Bundy

-0.63

 HOME

-0.62

 Administ

-0.62

 trivia

-0.61

perty

-0.60

Mex

-0.59

POSITIVE LOGITS

pole

0.83

obl

0.73

jamin

0.72

Whe

0.72

inite

0.70

steel

0.70

motion

0.68

dri

0.66

Medium

0.64

enna

0.63

Activations Density 0.334%

phrases referring to directions and positions

terms related to flatness and dimensionality in objects or concepts

No Comments

No Known Activations