INDEX

Explanations

significant nouns and terms related to institutions or authority

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 Whilst

-0.18

–

-0.18

 whilst

-0.17

–↵↵

-0.17

ienes

-0.15

________________________________________________________________

-0.14

fen

-0.14

kits

-0.14

phyl

-0.14

 anyways

-0.13

POSITIVE LOGITS

æĹĹ

0.18

 Pharma

0.17

 invent

0.16

 hoch

0.14

amping

0.14

 trillion

0.14

PRI

0.14

ÏĢÏģÎ¿

0.14

\@

0.14

 pretended

0.14

Activations Density 0.000%

significant nouns and terms related to institutions or authority

No Comments

No Known Activations