INDEX

Explanations

names and specific terms in multiple different languages

oai_token-act-pair · gpt-3.5-turbo

punctuation marks and related structures in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ModLoader

-0.87

avorite

-0.86

ãĥ¼ãĥĨãĤ£

-0.81

 glim

-0.79

backer

-0.77

tarian

-0.73

 superpower

-0.72

istries

-0.72

alogue

-0.70

emort

-0.70

POSITIVE LOGITS

 Amen

0.79

vim

0.73

 Levant

0.70

cÃ©

0.69

tu

0.69

tu

0.69

please

0.68

Tu

0.68

translation

0.68

 Indra

0.67

Activations Density 0.191%

names and specific terms in multiple different languages

punctuation marks and related structures in the text

No Comments

No Known Activations