INDEX

Explanations

instances of the phrase "how to" followed by verbs that suggest guidance or instruction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

anca

-0.18

ctions

-0.15

antly

-0.15

iliz

-0.15

 fewer

-0.14

patch

-0.14

OfWork

-0.14

ìķħ

-0.14

IDL

-0.14

patches

-0.13

POSITIVE LOGITS

0.15

assin

0.15

oping

0.15

obus

0.14

borne

0.14

MODE

0.14

ohana

0.13

acob

0.13

Rue

0.13

iba

0.12

Activations Density 0.043%

instances of the phrase "how to" followed by verbs that suggest guidance or instruction

No Comments

No Known Activations