INDEX

Explanations

references to pockets in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

sen

-0.18

zeitig

-0.17

eng

-0.17

sm

-0.16

si

-0.16

sb

-0.15

san

-0.15

yg

-0.15

Uy

-0.15

sz

-0.15

POSITIVE LOGITS

rung

0.19

omial

0.17

itan

0.16

pit

0.16

ized

0.16

ization

0.16

laus

0.16

atrix

0.16

ishment

0.15

loub

0.15

Activations Density 0.063%

references to pockets in various contexts

No Comments

No Known Activations