INDEX

Explanations

references to behaviors and behavioral characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

xing

-0.17

ãĥĥãĤ¯ãĤ¹

-0.16

owie

-0.16

ultip

-0.15

endale

-0.15

ICATION

-0.15

apus

-0.14

amet

-0.14

elo

-0.14

åĪĩãĤĬ

-0.14

POSITIVE LOGITS

fully

0.17

overe

0.16

 trop

0.16

 behaviour

0.15

uated

0.15

uelle

0.14

 patterns

0.14

acz

0.14

 behavior

0.14

amp

0.14

Activations Density 0.047%

references to behaviors and behavioral characteristics

No Comments

No Known Activations