INDEX

Explanations

phrases indicating the act of leaving an impact or impression

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

inkel

-0.19

 éł

-0.15

ureka

-0.15

Shield

-0.15

 termin

-0.13

verter

-0.13

rote

-0.13

_cpus

-0.13

POSITIVE LOGITS

 traces

0.47

 footprint

0.45

 marks

0.44

 trace

0.44

 mark

0.42

 imprint

0.42

 tracks

0.41

 prints

0.39

 foot

0.37

marks

0.37

Activations Density 0.196%

phrases indicating the act of leaving an impact or impression

No Comments

No Known Activations