INDEX

Explanations

references to data displayed in figures and tables

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

auer

-0.18

afa

-0.14

eprom

-0.13

Ã¼len

-0.13

pton

-0.13

malink

-0.13

:UIAlert

-0.13

į¼

-0.13

thon

-0.12

dae

-0.12

POSITIVE LOGITS

 figure

1.04

 Figure

0.98

Fig

0.90

fig

0.86

figure

0.84

 figura

0.82

-figure

0.76

Figure

0.75

Fig

0.75

fig

0.74

Activations Density 0.600%

references to data displayed in figures and tables

No Comments

No Known Activations