INDEX

Explanations

phrases related to universities and specific institutions

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to specific organizations, universities, and individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

WER

-0.75

edIn

-0.75

efeated

-0.70

rock

-0.70

ifies

-0.67

 Squirrel

-0.67

anamo

-0.66

alos

-0.64

chant

-0.64

 Geral

-0.64

POSITIVE LOGITS

ĸļ

1.05

coni

0.95

 Parkway

0.84

 Vanderbilt

0.84

rolet

0.76

owell

0.73

utic

0.72

awar

0.72

heimer

0.72

aternity

0.70

Activations Density 0.026%

phrases related to universities and specific institutions

proper nouns related to specific organizations, universities, and individuals

No Comments

No Known Activations