INDEX

Explanations

instances of the word "ed," particularly in reference to actions or practices that have been completed or designated

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

antar

-0.15

aul

-0.15

assa

-0.14

kul

-0.14

zo

-0.14

 permanent

-0.14

ega

-0.14

opp

-0.13

cash

-0.13

-feed

-0.13

POSITIVE LOGITS

WEEN

0.15

 Trafford

0.15

},{↵

0.14

IODevice

0.14

}s

0.14

ByExample

0.14

tera

0.14

lica

0.14

oldem

0.14

ship

0.13

Activations Density 0.012%

instances of the word "ed," particularly in reference to actions or practices that have been completed or designated

No Comments

No Known Activations