INDEX

Explanations

phrases containing the word "Trump."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Trump" in various contexts

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

âĸ¬

-0.77

ĪĴ

-0.72

 ende

-0.66

VERTISEMENT

-0.65

é¾įå

-0.64

YA

-0.64

FY

-0.62

vironment

-0.62

jab

-0.62

WithNo

-0.62

POSITIVE LOGITS

etsk

1.05

osit

0.99

olit

0.85

oleon

0.81

oly

0.80

olitan

0.79

ression

0.78

olicy

0.78

onent

0.78

ulse

0.76

Activations Density 0.016%

phrases containing the word "Trump."

occurrences of the word "Trump" in various contexts

No Comments

No Known Activations