INDEX

Explanations

details related to organizational tasks and scheduling

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

 scept

-0.18

Alright

-0.17

 honour

-0.16

 exposition

-0.16

 neighbourhood

-0.15

 honoured

-0.15

Behaviour

-0.15

cbc

-0.14

 EXEMPLARY

-0.14

wat

-0.14

POSITIVE LOGITS

 util

0.21

 which

0.18

 ÙĪØ§ÙĦØªÙĬ

0.17

which

0.17

 meaning

0.16

ÏģÎ³

0.16

å¾

0.16

 ÙĪØ°ÙĦÙĥ

0.15

taire

0.14

eg

0.14

Activations Density 0.657%

details related to organizational tasks and scheduling

No Comments

No Known Activations