INDEX

Explanations

random characters and seemingly unrelated words, possibly due to noise or errors in the data

oai_token-act-pair · gpt-3.5-turbo

unique or unusual characters and symbols

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

anwhile

-0.63

staking

-0.58

 behavi

-0.55

theless

-0.51

lest

-0.51

 agre

-0.50

vertisement

-0.49

hovah

-0.49

 compromises

-0.48

 streng

-0.47

POSITIVE LOGITS

ihara

0.62

pic

0.57

ii

0.56

âĢİ

0.56

çļĦ

0.55

ensis

0.53

rt

0.51

__

0.50

,[

0.49

()

0.48

Activations Density 0.399%

random characters and seemingly unrelated words, possibly due to noise or errors in the data

unique or unusual characters and symbols

No Comments

No Known Activations