INDEX

Explanations

words or phrases related to sequencing or step-by-step instructions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "then" indicating a sequence or transition in events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Latin

-0.69

MpServer

-0.68

toe

-0.67

 wart

-0.66

idad

-0.64

 Kane

-0.63

borgh

-0.63

 Eden

-0.61

assed

-0.60

undo

-0.60

POSITIVE LOGITS

 proceeded

1.08

©¶æ

0.94

 conclud

0.88

)</

0.84

 secondly

0.83

 proceed

0.81

 proceeds

0.81

 exha

0.80

etheless

0.79

Ń·

0.79

Activations Density 0.046%

words or phrases related to sequencing or step-by-step instructions

occurrences of the word "then" indicating a sequence or transition in events

No Comments

No Known Activations