INDEX

Explanations

words related to a slight or minor degree of something

oai_token-act-pair · gpt-3.5-turbo

references to minor nuances or subtle effects

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

mberg

-0.80

CDC

-0.71

halla

-0.67

 Ready

-0.67

clamation

-0.65

ceans

-0.64

ILLE

-0.63

yrs

-0.62

 Gutenberg

-0.62

tested

-0.61

POSITIVE LOGITS

 slight

0.81

 exaggeration

0.81

 resemblance

0.77

 modification

0.77

azaki

0.77

nesses

0.76

 easing

0.76

rance

0.75

 annoyance

0.74

 sway

0.74

Activations Density 0.012%

words related to a slight or minor degree of something

references to minor nuances or subtle effects

No Comments

No Known Activations