INDEX

Explanations

instances where the term "users" is explicitly mentioned

oai_token-act-pair · gpt-3.5-turbo

references to users in various contexts within the text

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

 Baptist

-0.69

amer

-0.68

western

-0.68

SourceFile

-0.65

 Maid

-0.64

 UNESCO

-0.63

Rib

-0.61

erella

-0.60

tein

-0.60

Reloaded

-0.60

POSITIVE LOGITS

pace

1.13

hip

1.08

cript

0.94

 interface

0.92

 interfaces

0.91

ettings

0.89

paces

0.87

interface

0.86

hare

0.84

itaire

0.80

Activations Density 0.027%

instances where the term "users" is explicitly mentioned

references to users in various contexts within the text

No Comments

No Known Activations