INDEX

Explanations

references related to political figures and their actions

oai_token-act-pair · gpt-3.5-turbo

statements and claims regarding political figures and their positions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

}.

-0.57

allery

-0.55

enment

-0.54

'.

-0.53

iverse

-0.51

.$

-0.51

 respectively

-0.50

inis

-0.50

ousand

-0.49

lished

-0.49

POSITIVE LOGITS

"[

0.58

Crime

0.53

ratom

0.53

"'

0.53

 "âĢ¦

0.51

vind

0.51

"))

0.51

 snipers

0.50

 intentionally

0.50

NetMessage

0.50

Activations Density 3.313%

references related to political figures and their actions

statements and claims regarding political figures and their positions

No Comments

No Known Activations