INDEX

Explanations

negation terms or phrases indicating the absence of something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ono

-0.16

eil

-0.15

isco

-0.15

ãĥ¼ãĥ¬

-0.15

ISCO

-0.14

Ð²Ð¸Ñĩ

-0.14

elic

-0.14

REFERRED

-0.14

ê±

-0.14

agr

-0.14

POSITIVE LOGITS

anje

0.17

mo

0.17

 actual

0.15

Mo

0.15

 Daniel

0.15

MOT

0.15

 already

0.15

 else

0.14

ensch

0.14

Moj

0.14

Activations Density 0.009%

negation terms or phrases indicating the absence of something

No Comments

No Known Activations