INDEX

Explanations

political terms and entities

oai_token-act-pair · gpt-3.5-turbo

references to formal roles or positions within organizations

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

]."

-0.65

$.

-0.64

 thereof

-0.58

}.

-0.58

 thereto

-0.54

."

-0.54

)."

-0.52

 respectively

-0.52

.).

-0.51

."[

-0.51

POSITIVE LOGITS

 FANTASY

0.50

 Profile

0.48

 partName

0.46

resa

0.45

 GOODMAN

0.45

 âĢº

0.45

:=

0.44

odore

0.44

 nutshell

0.43

 Background

0.42

Activations Density 4.032%

political terms and entities

references to formal roles or positions within organizations

No Comments

No Known Activations