INDEX

Explanations

code-related terms and structures in programming and markup languages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

.java

-0.15

Ð¾Ð²Ð°Ð»

-0.15

atura

-0.15

GenerationStrategy

-0.15

sty

-0.14

avin

-0.14

å®¢

-0.14

phin

-0.14

uff

-0.14

POSITIVE LOGITS

 tolua

0.21

tolua

0.16

lua

0.15

ayah

0.15

.lua

0.15

skin

0.15

Lua

0.15

unk

0.15

 diffuse

0.15

 Skin

0.15

Activations Density 0.036%

code-related terms and structures in programming and markup languages

No Comments

No Known Activations