INDEX

Explanations

year references expressed in a four-digit format

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

izens

-0.18

theid

-0.16

ules

-0.15

eman

-0.15

tes

-0.15

positories

-0.15

thes

-0.15

dej

-0.15

edback

-0.15

POSITIVE LOGITS

0.19

0.18

0.17

bolt

0.17

st

0.16

acci

0.16

 Surg

0.15

jourd

0.15

Activations Density 0.142%

year references expressed in a four-digit format

No Comments

No Known Activations