INDEX

Explanations

elements that signify past experiences and comparisons to previous situations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hani

-0.15

aÅŁÄ±

-0.14

orre

-0.13

hiba

-0.13

eci

-0.13

oday

-0.13

 tomorrow

-0.12

Î³Ïĩ

-0.12

Ø§ÙĬØ¯

-0.12

 JFactory

-0.12

POSITIVE LOGITS

 previous

1.24

 prior

0.98

previous

0.97

 Previous

0.94

Previous

0.92

 earlier

0.89

 past

0.79

.previous

0.79

_previous

0.76

prior

0.76

Activations Density 0.816%

elements that signify past experiences and comparisons to previous situations

No Comments

No Known Activations