INDEX

Explanations

the word "Ļ" at high activations

oai_token-act-pair · gpt-3.5-turbo

occurrences of the character "Ļ" or its variations, indicating a focus on specific character strings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ignition

-0.75

 flowering

-0.74

 scattering

-0.74

 chunks

-0.71

 Annex

-0.65

 fodder

-0.65

 flourishing

-0.64

 drying

-0.64

 succession

-0.64

 diminishing

-0.63

POSITIVE LOGITS

ï¸ı

1.16

mand

1.13

agree

1.05

mir

0.90

should

0.89

know

0.89

selves

0.87

own

0.86

exist

0.86

ould

0.85

Activations Density 0.113%

the word "Ļ" at high activations

occurrences of the character "Ļ" or its variations, indicating a focus on specific character strings

No Comments

No Known Activations