INDEX

Explanations

programming-related comments and annotations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ue

-0.14

ro

-0.14

iat

-0.14

 ters

-0.14

drv

-0.13

 Farrell

-0.13

:param

-0.13

ist

-0.13

lit

-0.13

ator

-0.13

POSITIVE LOGITS

throws

0.17

onica

0.16

-fontawesome

0.15

orden

0.15

 Throws

0.15

 throws

0.14

asje

0.14

 thrown

0.14

 throwing

0.14

Ã´n

0.14

Activations Density 0.011%

programming-related comments and annotations

No Comments

No Known Activations