INDEX

Explanations

phrases that indicate a timeframe or duration

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

/goto

-0.15

à¥Ĥà¤¬à¤°

-0.15

vern

-0.15

postalcode

-0.15

 premi

-0.14

Î³ÏīÎ½

-0.14

asurer

-0.14

ìľ¡

-0.14

 Hanson

-0.14

zej

-0.14

POSITIVE LOGITS

ned

0.21

ning

0.16

izu

0.16

amb

0.15

 receptive

0.15

ãĥĥãĤ·ãĥ¥

0.15

æ¨ª

0.15

atic

0.15

iddleware

0.15

Activations Density 0.010%

phrases that indicate a timeframe or duration

No Comments

No Known Activations