INDEX

Explanations

concepts related to influence and impact

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

rong

-0.17

uele

-0.15

ngr

-0.15

rans

-0.14

abad

-0.14

amber

-0.14

 Particip

-0.14

lou

-0.14

 deem

-0.14

awa

-0.13

POSITIVE LOGITS

 bring

0.21

 brings

0.18

inf

0.18

 bringing

0.16

indr

0.16

aped

0.16

 Bring

0.16

-INF

0.15

 create

0.15

 prod

0.15

Activations Density 0.133%

concepts related to influence and impact

No Comments

No Known Activations