INDEX

Explanations

words related to contradiction or opposing viewpoints

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iliz

-0.18

å¡ļ

-0.17

icide

-0.16

uba

-0.15

ãĥ³ãĤ¯

-0.15

istrat

-0.14

scribe

-0.14

ILA

-0.14

ificance

-0.14

im

-0.14

POSITIVE LOGITS

 contr

0.30

 CONTR

0.24

 Contr

0.21

ictory

0.21

Contr

0.20

contr

0.19

ition

0.18

ived

0.17

actions

0.17

Activations Density 0.009%

words related to contradiction or opposing viewpoints

No Comments

No Known Activations