INDEX

Explanations

technical details and instructions related to data processing

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

peq

-0.16

agen

-0.15

Winvalid

-0.15

ipay

-0.14

_mE

-0.14

ÙĬÙĦØ§

-0.14

iana

-0.14

okoj

-0.14

jure

-0.13

ÏįÏĦÎµ

-0.13

POSITIVE LOGITS

 corresponding

0.29

 matching

0.26

 Correspond

0.24

 correspondent

0.24

 corres

0.23

 Matching

0.22

matching

0.22

 matched

0.21

 correspond

0.21

Matching

0.21

Activations Density 0.079%

technical details and instructions related to data processing

No Comments

No Known Activations