INDEX

Explanations

programming-related keywords, especially those related to visibility and access modifiers

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

inton

-0.19

cox

-0.17

BOVE

-0.15

sumer

-0.14

resident

-0.14

çĭĲ

-0.14

ãĤ«ãĥ¼

-0.14

Ø²Ùĩ

-0.14

 peanut

-0.14

enderror

-0.14

POSITIVE LOGITS

ante

0.17

jam

0.16

cape

0.15

issing

0.15

-HT

0.14

omor

0.14

aje

0.14

oro

0.13

Writable

0.13

FolderPath

0.13

Activations Density 0.007%

programming-related keywords, especially those related to visibility and access modifiers

No Comments

No Known Activations