INDEX

Explanations

instances of verbs ending in 'ing', particularly those related to actions or processes

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

sian

-0.18

shan

-0.17

sch

-0.17

screen

-0.16

anova

-0.15

Ã©gor

-0.15

ceipt

-0.15

suz

-0.15

sse

-0.14

smith

-0.14

POSITIVE LOGITS

redient

0.22

ton

0.21

enuity

0.21

haus

0.21

hausen

0.20

redients

0.19

ØªÙĪÙĨ

0.18

elli

0.18

REDIENT

0.17

ope

0.17

Activations Density 0.110%

instances of verbs ending in 'ing', particularly those related to actions or processes

No Known Activations