INDEX

Explanations

elements related to location and descriptors within a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 Scottish

-0.16

 Scotch

-0.15

olo

-0.15

ahas

-0.15

 knull

-0.15

qus

-0.15

olan

-0.15

Median

-0.14

median

-0.14

oldown

-0.14

POSITIVE LOGITS

ken

0.23

 guid

0.19

erra

0.17

 vera

0.17

richt

0.17

ae

0.17

wi

0.17

wis

0.16

sel

0.15

 efter

0.15

Activations Density 0.002%

elements related to location and descriptors within a narrative context

No Comments

No Known Activations