INDEX

Explanations

conjunctions and phrases that link ideas or clauses together

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

annies

-0.16

apolis

-0.15

mans

-0.14

elon

-0.14

Swe

-0.14

/fixtures

-0.14

 _______,

-0.13

enson

-0.13

mlink

-0.13

nze

-0.13

POSITIVE LOGITS

 finally

0.21

finally

0.17

 Lastly

0.16

Lastly

0.16

lick

0.15

Finally

0.14

 Finally

0.14

hic

0.14

enek

0.14

licts

0.13

Activations Density 0.100%

conjunctions and phrases that link ideas or clauses together

No Comments

No Known Activations