INDEX

Explanations

occurrences of mathematical symbols and expressions involving combinatorial concepts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Stick

-0.17

 Evel

-0.16

 stick

-0.15

-0.14

 aerial

-0.14

 Pied

-0.14

ãģĸ

-0.14

LOOP

-0.14

estate

-0.14

.getZ

-0.14

POSITIVE LOGITS

atrib

0.17

kits

0.15

stown

0.15

/start

0.15

Å¡Å¥

0.14

_cmos

0.14

 lÃłnh

0.14

erson

0.14

ups

0.14

Activations Density 0.007%

occurrences of mathematical symbols and expressions involving combinatorial concepts

No Comments

No Known Activations