INDEX

Explanations

occurrences of the word "to."

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

AKER

-0.16

AU

-0.14

coe

-0.14

gel

-0.14

Ãµes

-0.14

 wearing

-0.14

unb

-0.14

gre

-0.14

Gel

-0.14

 Kraj

-0.14

POSITIVE LOGITS

urette

0.15

readcr

0.15

.timezone

0.15

é³

0.14

èĮ¨

0.14

avana

0.14

inue

0.14

sert

0.14

Mine

0.14

 Boeh

0.13

Activations Density 0.025%

occurrences of the word "to."

No Comments

No Known Activations