INDEX

Explanations

phrases that indicate actions, conditions, or obligations

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

iron

-0.16

èĪĴ

-0.16

TargetException

-0.15

 Beste

-0.15

rosse

-0.15

 má»Ł

-0.14

adar

-0.14

åľ³

-0.14

orate

-0.14

.Axis

-0.14

POSITIVE LOGITS

ajan

0.14

quist

0.14

rod

0.14

plit

0.14

uster

0.13

bonus

0.13

asal

0.13

udad

0.13

eon

0.13

-proxy

0.13

Activations Density 0.002%

phrases that indicate actions, conditions, or obligations

No Comments

No Known Activations