INDEX

Explanations

definite articles and conjunctions indicating connection or equivalence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

uml

-0.15

 nhau

-0.14

ÙĭØ§

-0.14

 Rates

-0.14

uve

-0.14

ialis

-0.14

UIT

-0.14

.this

-0.14

.PERMISSION

-0.14

jem

-0.14

POSITIVE LOGITS

ar

0.15

0.14

offs

0.14

py

0.14

Å¯

0.14

ville

0.14

reater

0.14

 Moderator

0.13

 Isaac

0.13

Activations Density 0.024%

definite articles and conjunctions indicating connection or equivalence

No Comments

No Known Activations