INDEX

Explanations

mentions of specific entities or locations involved in geopolitical events

oai_token-act-pair · gpt-3.5-turbo

references to organizations, places, and significant themes associated with current events and socio-political issues

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

.).

-0.61

CLASSIFIED

-0.60

Reloaded

-0.54

").

-0.52

agra

-0.52

]."

-0.51

).[

-0.51

".[

-0.51

irlf

-0.50

)).

-0.50

POSITIVE LOGITS

 varies

0.61

isphere

0.60

 meanwhile

0.57

 arises

0.57

 coincided

0.55

 differed

0.54

iens

0.53

wealth

0.53

 depends

0.52

 relates

0.51

Activations Density 1.508%

mentions of specific entities or locations involved in geopolitical events

references to organizations, places, and significant themes associated with current events and socio-political issues

No Comments

No Known Activations