INDEX

Explanations

occurrences of the word "continuing" or similar forms indicating ongoing processes or activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

lesc

-0.19

agnost

-0.16

hiro

-0.16

arme

-0.15

strate

-0.15

borg

-0.15

licht

-0.14

ãĥīãĥ«

-0.14

ehler

-0.14

bersome

-0.14

POSITIVE LOGITS

ĶåĽŀ

0.16

onus

0.16

imer

0.15

åīĤ

0.15

pons

0.15

ons

0.14

linky

0.14

AdapterFactory

0.14

itchen

0.14

rophe

0.14

Activations Density 0.006%

occurrences of the word "continuing" or similar forms indicating ongoing processes or activities

No Comments

No Known Activations