INDEX

Explanations

programming-related keywords and functions in code snippets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

otec

-0.15

etal

-0.14

rastructure

-0.14

çĽĺ

-0.13

odash

-0.13

ORTH

-0.13

cea

-0.13

alÄ±k

-0.13

utta

-0.13

ifton

-0.13

POSITIVE LOGITS

alc

0.21

cade

0.17

reeze

0.16

errupted

0.15

ãĥĥãĥģ

0.14

âĨĵ

0.14

 Haram

0.14

 âĨĵ

0.14

æ£ĭ

0.14

iali

0.14

Activations Density 0.059%

programming-related keywords and functions in code snippets

No Comments

No Known Activations