INDEX

Explanations

the word "into" and its variations in various contexts, often related to transition or transformation

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ì¡±

-0.18

iller

-0.18

ened

-0.16

ified

-0.15

liest

-0.14

wyn

-0.14

 undermin

-0.14

UsageId

-0.14

enson

-0.14

lint

-0.14

POSITIVE LOGITS

/out

0.33

/from

0.20

 obliv

0.20

 orbit

0.18

yre

0.18

 submission

0.17

 existence

0.17

 action

0.17

-the

0.16

Activations Density 0.102%

the word "into" and its variations in various contexts, often related to transition or transformation

No Comments

No Known Activations