INDEX

Explanations

references to seeds or seed-related activities

oai_token-act-pair · gpt-3.5-turbo

references to seeds

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

paio

-0.81

urches

-0.81

pload

-0.79

cedented

-0.78

olulu

-0.77

quished

-0.77

 mathemat

-0.73

vernment

-0.73

ufact

-0.71

inez

-0.71

POSITIVE LOGITS

lings

1.09

seed

1.07

 Seeds

1.00

 seed

0.94

icons

0.89

ãĥ¼ãĥĨ

0.88

 seeds

0.87

ling

0.85

less

0.82

orf

0.81

Activations Density 0.013%

references to seeds or seed-related activities

references to seeds

No Comments

No Known Activations