INDEX

Explanations

instances of people and their actions or states of being

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 æĭ

-0.17

 reinterpret

-0.15

asil

-0.15

izza

-0.15

enant

-0.15

angl

-0.15

ennon

-0.14

ombs

-0.14

.pp

-0.14

aware

-0.13

POSITIVE LOGITS

 seen

0.36

seen

0.33

 Seen

0.33

heard

0.32

 heard

0.32

 spotted

0.31

Seen

0.31

 Heard

0.28

_seen

0.25

 observed

0.24

Activations Density 0.037%

instances of people and their actions or states of being

No Comments

No Known Activations