INDEX

Explanations

references to openings, access, or breaking in contexts of various scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

cdn

-0.17

umba

-0.15

Äįky

-0.15

spath

-0.15

/*******************************************************************************↵

-0.15

rim

-0.14

canf

-0.14

Ð½Ð°Ð¼

-0.14

 ÙĨÙģ

-0.14

cgi

-0.14

POSITIVE LOGITS

lust

0.16

Ø³Ø§ÙĨÛĮ

0.16

 Weiner

0.15

aille

0.14

 single

0.14

 Rever

0.14

urum

0.14

assin

0.13

ml

0.13

Activations Density 0.280%

references to openings, access, or breaking in contexts of various scenarios

No Comments

No Known Activations