INDEX

Explanations

references to time periods and comparisons over different durations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

Î¸Î¹

-0.18

ç°

-0.14

ãĤĩ

-0.14

ander

-0.14

å¼ı

-0.14

riger

-0.14

agli

-0.14

rdf

-0.14

 early

-0.13

eah

-0.13

POSITIVE LOGITS

 period

0.26

 periods

0.19

 periodo

0.19

visor

0.19

/autoload

0.18

_period

0.18

Period

0.18

.period

0.17

_periods

0.17

period

0.17

Activations Density 0.079%

references to time periods and comparisons over different durations

No Comments

No Known Activations