INDEX

Explanations

phrases related to the positioning or orientation of objects

oai_token-act-pair · gpt-3.5-turbo

phrases related to the concept of "of" indicating relationships or ownerships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

hran

-0.84

©¶æ¥µ

-0.78

umerable

-0.77

nos

-0.76

arak

-0.75

center

-0.74

hra

-0.73

\/\/

-0.72

=""

-0.71

swick

-0.71

POSITIVE LOGITS

 incoming

0.85

 each

0.85

 these

0.83

 those

0.78

 Nanto

0.78

 your

0.68

 course

0.68

the

0.67

our

0.65

 metals

0.65

Activations Density 0.163%

phrases related to the positioning or orientation of objects

phrases related to the concept of "of" indicating relationships or ownerships

No Comments

No Known Activations