INDEX

Explanations

programming-related identifiers and functions within code segments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

gunakan

-0.17

ancer

-0.14

allee

-0.13

HLT

-0.13

.writeln

-0.13

flu

-0.13

 Sterling

-0.13

usaha

-0.13

unei

-0.13

 Saga

-0.13

POSITIVE LOGITS

 Ð´Ð¾ÑĢ

0.14

\Extension

0.14

hek

0.14

pecified

0.14

emma

0.14

Î¼ÏĢÎ¿

0.13

BIT

0.13

iling

0.13

ainen

0.13

ITH

0.13

Activations Density 0.228%

programming-related identifiers and functions within code segments

No Comments

No Known Activations