INDEX

Explanations

programming-related terminology, specifically in the context of functions and user-related objects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

alse

-0.18

arend

-0.17

al

-0.16

-0.15

ach

-0.15

æĢģ

-0.15

kee

-0.15

kut

-0.15

Ã¡

-0.14

bit

-0.14

POSITIVE LOGITS

 Getter

0.19

ssel

0.19

 getter

0.19

#get

0.15

 getenv

0.15

 getters

0.15

inline

0.15

 Camb

0.15

_inline

0.15

İ·

0.15

Activations Density 0.025%

programming-related terminology, specifically in the context of functions and user-related objects

No Comments

No Known Activations