INDEX

Explanations

terms related to conditions or attributes that imply comparison or limitations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

illac

-0.18

vertisement

-0.17

ToObject

-0.16

presso

-0.16

onium

-0.15

TickCount

-0.15

enne

-0.14

ISCO

-0.14

payer

-0.14

WARN

-0.14

POSITIVE LOGITS

 becomes

0.18

 became

0.16

æĺ¯ä¸Ģä¸ª

0.16

 becoming

0.16

 become

0.15

bec

0.15

 Being

0.14

 Became

0.14

 being

0.14

andal

0.14

Activations Density 0.003%

terms related to conditions or attributes that imply comparison or limitations

No Comments

No Known Activations