INDEX

Explanations

statements related to significant events or actions that could be interpreted as politically controversial or impactful

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

Ø§ÙĨÛĮ

-0.16

Kad

-0.16

ami

-0.15

owel

-0.14

aea

-0.14

ow

-0.14

ÙħØ§

-0.14

oty

-0.14

%A

-0.14

opleft

-0.14

POSITIVE LOGITS

Ã©rc

0.17

Ø´ÙĪ

0.15

moth

0.15

lobber

0.15

 Lehr

0.15

pez

0.15

à¤¤à¤®

0.14

esson

0.14

ortho

0.14

ighton

0.14

Activations Density 0.029%

statements related to significant events or actions that could be interpreted as politically controversial or impactful

No Comments

No Known Activations