INDEX

Explanations

contexts related to teaching and language learning

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ê

-0.15

 Fucking

-0.15

.SizeType

-0.14

uri

-0.13

Ä±s

-0.13

ioni

-0.13

imar

-0.12

ako

-0.12

erno

-0.12

POSITIVE LOGITS

ÃħÅ¸

0.18

 Ã¢

0.15

ÃĤ

0.14

 cÃ¢

0.14

REATED

0.14

ÃĦÅ¸

0.14

 aforementioned

0.14

_tac

0.14

 Ãİ

0.14

Ã¢b

0.14

Activations Density 0.315%

contexts related to teaching and language learning

No Comments

No Known Activations