INDEX

Explanations

actions and outcomes related to winning and losing in competitive contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ê

-0.16

lauf

-0.15

LOOP

-0.15

ç©´

-0.14

åı·

-0.14

habit

-0.14

rette

-0.14

 Explanation

-0.14

é»ĺ

-0.14

_KERNEL

-0.13

POSITIVE LOGITS

azar

0.16

reek

0.16

æĪ¸

0.14

/LICENSE

0.13

itk

0.13

rese

0.13

istrovstvÃŃ

0.13

nie

0.13

ng

0.13

oste

0.13

Activations Density 0.062%

actions and outcomes related to winning and losing in competitive contexts

No Comments

No Known Activations