INDEX

Explanations

references to specific locations or geographical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

åĮ

-0.16

oner

-0.14

angling

-0.14

ãĥĥãĤ·ãĥ¥

-0.14

isch

-0.14

 Farr

-0.14

 comple

-0.13

aÃ§

-0.13

sealed

-0.13

artner

-0.13

POSITIVE LOGITS

Ãºmer

0.15

 cÃ©lib

0.14

 adulti

0.14

resas

0.14

GIN

0.14

 adultes

0.14

ãģĹãģ®

0.13

ilenames

0.13

ples

0.13

 oppon

0.13

Activations Density 0.023%

references to specific locations or geographical contexts

No Comments

No Known Activations