INDEX

Explanations

references to personal blogs and the act of blogging itself

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

upal

-0.16

ixel

-0.16

agate

-0.15

 Flesh

-0.14

aviest

-0.14

ister

-0.14

habi

-0.14

ucken

-0.14

ulace

-0.14

ube

-0.13

POSITIVE LOGITS

mus

0.17

arella

0.16

ROOT

0.14

ged

0.14

div

0.14

-mf

0.14

quia

0.14

conditionally

0.14

 sarc

0.14

/apt

0.14

Activations Density 0.099%

references to personal blogs and the act of blogging itself

No Comments

No Known Activations