INDEX

Explanations

specific formatting directives and physical attributes related to documents and text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 trunc

-0.17

 visto

-0.16

align

-0.16

 Lost

-0.16

 anch

-0.16

 taken

-0.16

 dispatch

-0.15

lace

-0.15

Made

-0.15

 implant

-0.15

POSITIVE LOGITS

iced

0.19

 armored

0.17

 armed

0.17

 denn

0.17

enced

0.17

 themed

0.17

 haired

0.16

 gated

0.16

LATED

0.16

 boarded

0.16

Activations Density 0.431%

specific formatting directives and physical attributes related to documents and text

No Comments

No Known Activations