INDEX

Explanations

phrases emphasizing qualitative descriptions, particularly adjectives and adverbs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

rame

-0.15

us

-0.15

bak

-0.15

 Ø¯ÙĦ

-0.15

them

-0.14

ez

-0.14

Ä

-0.14

oints

-0.14

oke

-0.14

POSITIVE LOGITS

saldo

0.15

pty

0.15

alyzed

0.15

 duct

0.15

erglass

0.14

 CreateMap

0.14

è°±

0.14

.Doc

0.14

Ø¬ÙĬÙĦ

0.14

orthand

0.13

Activations Density 0.019%

phrases emphasizing qualitative descriptions, particularly adjectives and adverbs

No Comments

No Known Activations