INDEX

Explanations

texts related to news events

oai_token-act-pair · gpt-3.5-turbo

phrases related to incidents or events, particularly those involving emergencies or reactions

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

etheless

-0.81

zbollah

-0.79

 endeav

-0.75

 trivial

-0.74

 domestically

-0.72

cffff

-0.71

 estranged

-0.70

 trunc

-0.69

 lodged

-0.69

 fors

-0.69

POSITIVE LOGITS

Testing

1.07

Prof

1.06

Said

1.06

Contribut

1.05

Indeed

1.04

SPONSORED

1.04

Thirty

1.02

Another

0.99

Attempts

0.98

Added

0.98

Activations Density 0.226%

texts related to news events

phrases related to incidents or events, particularly those involving emergencies or reactions

No Comments

No Known Activations