INDEX

Explanations

phrases related to movies or TV shows

oai_token-act-pair · gpt-3.5-turbo

periods or punctuation marks at the end of sentences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 glim

-0.82

 affordable

-0.79

 answer

-0.77

 pesky

-0.76

 questioning

-0.76

 suspic

-0.76

 tremend

-0.75

plet

-0.75

 elusive

-0.75

 crunch

-0.74

POSITIVE LOGITS

 Additionally

1.51

 Alternatively

1.41

 However

1.37

 Also

1.32

 Later

1.31

 Afterwards

1.31

 Similarly

1.29

 Interestingly

1.26

 Likewise

1.22

 Previously

1.22

Activations Density 0.482%

phrases related to movies or TV shows

periods or punctuation marks at the end of sentences

No Comments

No Known Activations