INDEX

Explanations

programming and coding-related syntax or keywords

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ervlet

-0.17

LK

-0.15

 Abrams

-0.14

eczy

-0.14

olley

-0.14

nelle

-0.14

èĺ

-0.13

emmel

-0.13

odie

-0.13

ilit

-0.13

POSITIVE LOGITS

"struct

0.15

ãĥ£

0.14

ineTransform

0.14

rypton

0.14

vice

0.14

URN

0.14

 --------------------------------------------------------------------------↵

0.14

<quote

0.14

pcodes

0.13

å±ı

0.13

Activations Density 0.001%

programming and coding-related syntax or keywords

No Comments

No Known Activations