INDEX

Explanations

adjectives and their variations in detail

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

INGS

-0.18

ings

-0.16

Ã©nom

-0.16

neh

-0.15

ennon

-0.15

ulas

-0.15

morgan

-0.14

eprom

-0.14

Ownership

-0.13

iÃŁ

-0.13

POSITIVE LOGITS

unct

0.32

ourn

0.30

acency

0.26

acent

0.25

oint

0.24

unction

0.22

unt

0.21

UNCT

0.21

acente

0.21

oints

0.20

Activations Density 0.009%

adjectives and their variations in detail

No Comments

No Known Activations