INDEX

Explanations

words associated with the character 'Er' and its variations, often related to people's names or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

enko

-0.16

acre

-0.15

iliary

-0.15

Ð·Ð²Ð¸ÑĩÐ°Ð¹

-0.14

iques

-0.14

gf

-0.14

ylvania

-0.14

MASK

-0.14

-0.13

Ngh

-0.13

POSITIVE LOGITS

linger

0.19

asmus

0.18

upert

0.16

hyth

0.16

utan

0.15

hea

0.15

hythm

0.15

hardt

0.15

izabeth

0.15

yst

0.15

Activations Density 0.038%

words associated with the character 'Er' and its variations, often related to people's names or titles

No Comments

No Known Activations