INDEX

Explanations

patterns of similarity and repetition in contexts

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

deaux

-0.16

zap

-0.16

EFAULT

-0.15

ocs

-0.14

kb

-0.14

atos

-0.14

 Settlement

-0.14

ä¹Łä¸į

-0.14

-0.13

 Bounty

-0.13

POSITIVE LOGITS

ä¸Ģæł·

0.16

likle

0.14

iado

0.14

Î¯Î´Î±

0.14

geh

0.14

anj

0.14

 similarly

0.14

.ease

0.14

prech

0.14

oref

0.14

Activations Density 0.060%

patterns of similarity and repetition in contexts

No Comments

No Known Activations