INDEX

Explanations

phrases related to transformation or conversion of objects or concepts into something else

oai_token-act-pair · gpt-3.5-turbo

terms related to transformation or conversion processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pport

-0.63

 rall

-0.56

 conduc

-0.54

vertisement

-0.53

 Comparison

-0.51

venge

-0.51

emate

-0.50

 Warrant

-0.50

 reminder

-0.50

 Detail

-0.49

POSITIVE LOGITS

 into

2.44

into

2.19

 Into

1.99

 INTO

1.96

 onto

1.74

 seamlessly

0.97

 toward

0.82

 overboard

0.79

 towards

0.77

aneously

0.76

Activations Density 0.472%

phrases related to transformation or conversion of objects or concepts into something else

terms related to transformation or conversion processes

No Comments

No Known Activations