INDEX

Explanations

keywords related to a specific online news platform named "Infowars."

oai_token-act-pair · gpt-3.5-turbo

instances of the term "Inf" followed by various numbers or contexts suggesting information or reference points

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

BOOK

-0.80

âķĲâķĲ

-0.78

\\\\\\\\

-0.77

 Weir

-0.72

ãģ®éŃĶ

-0.68

 Athletic

-0.68

UAL

-0.68

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.68

manship

-0.67

OPLE

-0.66

POSITIVE LOGITS

rared

1.23

rastructure

1.20

inite

1.19

luence

1.12

ractions

1.11

amous

1.10

ertility

1.10

inity

1.09

iltration

1.05

lamm

1.04

Activations Density 0.013%

keywords related to a specific online news platform named "Infowars."

instances of the term "Inf" followed by various numbers or contexts suggesting information or reference points

No Comments

No Known Activations