INDEX

Explanations

references to penalties and awarded plays in sports contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

å¼¥

-0.16

æģ¯

-0.15

Va

-0.15

åħ¥ãĤĮ

-0.14

iges

-0.14

_AXIS

-0.14

ãģķãĤĵãģ®

-0.14

_TRACE

-0.13

Ups

-0.13

AGO

-0.13

POSITIVE LOGITS

oder

0.19

 Weld

0.16

issance

0.16

á»ģn

0.15

 Fucked

0.15

kea

0.15

oli

0.15

Ã¼stÃ¼

0.15

ancode

0.15

-HT

0.14

Activations Density 0.009%

references to penalties and awarded plays in sports contexts

No Comments

No Known Activations