INDEX

Explanations

documents or files that are attached or linked

oai_token-act-pair · gpt-3.5-turbo

words related to attachment or being connected

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ativity

-0.71

icky

-0.67

drm

-0.66

 deliveries

-0.65

states

-0.65

atively

-0.64

 shipments

-0.63

 easing

-0.63

 stocking

-0.62

appy

-0.62

POSITIVE LOGITS

 Thumbnails

0.77

anooga

0.77

sie

0.75

ached

0.75

achment

0.74

 Dame

0.70

edIn

0.69

 Swordsman

0.68

icut

0.67

 Aires

0.67

Activations Density 0.015%

documents or files that are attached or linked

words related to attachment or being connected

No Comments

No Known Activations