INDEX

Explanations

phrases or terms related to documentation and classification systems

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

-fontawesome

-0.15

iversit

-0.15

ased

-0.15

kke

-0.15

 Shelf

-0.14

ueil

-0.14

ÑģÐ¸Ð¾Ð½

-0.14

Ã¶s

-0.14

ÏĦÎ¯Î±

-0.14

ãĥªãĤ«

-0.14

POSITIVE LOGITS

Ø¯Ø§Ø±ÛĮ

0.16

quir

0.14

ãĥĭãĥ¼

0.14

ÑĥÐ¶Ðµ

0.14

kehr

0.13

autop

0.13

EncodingException

0.13

QUEST

0.13

iele

0.13

 Kirby

0.13

Activations Density 0.249%

phrases or terms related to documentation and classification systems

No Comments

No Known Activations