INDEX

Explanations

phrases indicating final thoughts or conclusions

oai_token-act-pair · gpt-3.5-turbo

sentence-ending punctuation

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 volunte

-0.87

ascus

-0.84

 exting

-0.83

 tremend

-0.79

 teasp

-0.79

 satell

-0.79

 carbohyd

-0.77

 suspic

-0.77

 pione

-0.74

 earthqu

-0.73

POSITIVE LOGITS

 Though

1.42

But

1.39

 While

1.32

 Whether

1.31

 According

1.30

 Nevertheless

1.30

 Nonetheless

1.30

 Although

1.29

 That

1.27

 However

1.26

Activations Density 0.560%

phrases indicating final thoughts or conclusions

sentence-ending punctuation

No Comments

No Known Activations