INDEX

Explanations

elements related to structure or organization within a text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

icas

-0.16

ledo

-0.15

iger

-0.15

icit

-0.14

 draped

-0.14

ovo

-0.14

rado

-0.14

SCAN

-0.14

sonian

-0.14

avad

-0.14

POSITIVE LOGITS

-pocket

0.17

IQ

0.15

¼

0.15

eshire

0.14

çŁ

0.14

 pocket

0.14

cola

0.14

 kalk

0.14

IQ

0.14

reeNode

0.13

Activations Density 0.002%

elements related to structure or organization within a text

No Comments

No Known Activations