INDEX

Explanations

phrases related to philosophical or abstract concepts, possibly including contrasting ideas

oai_token-act-pair · gpt-3.5-turbo

sentences or clauses that convey strong assertions or conclusions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 costing

-0.75

ascus

-0.75

abouts

-0.74

 teasp

-0.73

 nationwide

-0.73

eport

-0.73

 earthqu

-0.72

 numbered

-0.71

 wire

-0.71

 officially

-0.70

POSITIVE LOGITS

 Whereas

1.60

 Hence

1.47

 Often

1.40

 Thus

1.40

 Consequently

1.40

 Conversely

1.34

 Sometimes

1.31

 Likewise

1.31

 Whether

1.28

 Such

1.28

Activations Density 0.452%

phrases related to philosophical or abstract concepts, possibly including contrasting ideas

sentences or clauses that convey strong assertions or conclusions

No Comments

No Known Activations