INDEX

Explanations

names of political figures and terms related to political campaigns

oai_token-act-pair · gpt-3.5-turbo

comparisons of election outcomes and candidates' standings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

sterdam

-0.82

obyl

-0.77

Soc

-0.77

xxx

-0.76

abba

-0.76

 Rooms

-0.76

ĸļ

-0.73

pora

-0.73

arta

-0.72

nai

-0.71

POSITIVE LOGITS

 Hillary

1.44

 Democrat

1.43

 Barack

1.30

 frontrunner

1.28

 Romney

1.28

 incumbent

1.28

 challenger

1.27

 Bernie

1.27

 presumptive

1.27

 Clinton

1.25

Activations Density 0.234%

names of political figures and terms related to political campaigns

comparisons of election outcomes and candidates' standings

No Comments

No Known Activations