INDEX

Explanations

references to timeframes and durations

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

lfw

-0.16

uzzi

-0.15

ItemSelectedListener

-0.15

ersh

-0.15

StartTime

-0.14

*(*

-0.14

mey

-0.14

 Falk

-0.14

cheid

-0.14

_flutter

-0.14

POSITIVE LOGITS

 frame

0.43

frame

0.36

frames

0.32

Frame

0.32

FRAME

0.31

 Frame

0.31

-frame

0.31

 frames

0.30

 range

0.30

_FRAME

0.27

Activations Density 0.052%

references to timeframes and durations

No Comments

No Known Activations