INDEX

Explanations

pronouns 'I' followed by a verb

oai_token-act-pair · gpt-3.5-turbo

the usage of the pronoun "I"

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

tains

-0.72

pires

-0.63

 Philipp

-0.63

heads

-0.62

tnc

-0.61

 Mehran

-0.57

Gap

-0.56

 excess

-0.55

 indistinguishable

-0.54

 Reverse

-0.53

POSITIVE LOGITS

'm

1.34

've

1.28

'll

1.17

 suppose

1.04

'd

1.02

WI

1.01

nex

0.96

deals

0.93

 dunno

0.91

vor

0.89

Activations Density 0.206%

pronouns 'I' followed by a verb

the usage of the pronoun "I"

No Comments

No Known Activations