INDEX

Explanations

references to winning and losing 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 @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

ç¹ģ

-0.15

ottes

-0.15

asset

-0.15

ÄĽÅ¾

-0.15

compat

-0.15

apur

-0.14

lotte

-0.14

quired

-0.14

ilg

-0.14

//{{

-0.14

POSITIVE LOGITS

 execution

0.26

 Execution

0.24

 execute

0.21

 tonight

0.20

 Execute

0.20

execute

0.20

 executing

0.19

 executed

0.19

Execution

0.19

_execution

0.19

Activations Density 0.060%

references to winning and losing in sports contexts

No Comments

No Known Activations