INDEX

Explanations

references to time elapsed between events

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

kke

-0.15

 gÃ¶n

-0.14

antis

-0.14

Ø§ÙĤ

-0.14

ivating

-0.13

_GP

-0.13

erli

-0.13

Bab

-0.13

crate

-0.13

IMIT

-0.13

POSITIVE LOGITS

still

0.18

 still

0.18

rob

0.17

 STILL

0.17

 Still

0.17

ä»į

0.16

Still

0.15

 encore

0.15

Eff

0.15

Rob

0.15

Activations Density 0.043%

references to time elapsed between events

No Comments

No Known Activations