INDEX

Explanations

mentions of items or entities that are encompassed or contained within another entity or context

oai_token-act-pair · gpt-3.5-turbo

instances of the word "included."

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

aptic

-0.74

thur

-0.72

HCR

-0.70

redo

-0.70

ethy

-0.68

²¾

-0.67

CLASSIFIED

-0.67

ggles

-0.66

veyard

-0.65

mosp

-0.64

POSITIVE LOGITS

 prominently

0.78

 herein

0.70

ãģĨ

0.68

ãĤ´

0.64

ãĥĬ

0.63

ãĤ½

0.62

rescent

0.61

ropolitan

0.60

 Include

0.60

atinum

0.59

Activations Density 0.027%

mentions of items or entities that are encompassed or contained within another entity or context

instances of the word "included."

No Comments

No Known Activations