INDEX

Explanations

terms related to programming constructs and software functionality

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

hus

-0.19

.getRaw

-0.16

reich

-0.15

 æ³

-0.15

Ø§ÙĨÙĪÙĨ

-0.15

ripp

-0.14

covered

-0.14

uga

-0.14

 mari

-0.14

alker

-0.14

POSITIVE LOGITS

ering

0.15

lamaz

0.15

ä»ķ

0.15

gang

0.14

itti

0.14

llvm

0.14

oker

0.14

.dump

0.14

eters

0.14

interop

0.14

Activations Density 0.024%

terms related to programming constructs and software functionality

No Comments

No Known Activations