INDEX

Explanations

terms related to technical language and coding in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.16

 tricks

-0.16

miss

-0.16

utt

-0.16

ãĥĵãĥ¼

-0.15

BITS

-0.15

 enumer

-0.15

gem

-0.14

opper

-0.14

POSITIVE LOGITS

ecz

0.16

iyim

0.16

çłĤ

0.15

kses

0.15

SocketAddress

0.15

lopedia

0.15

ToJson

0.15

.Pod

0.15

ä¹ħä¹ħ

0.15

ngÃ¶r

0.14

Activations Density 0.013%

terms related to technical language and coding in various contexts

No Comments

No Known Activations