INDEX

Explanations

phrases related to proportion and quantity

oai_token-act-pair · gpt-3.5-turbo

terms related to scaling and adjustments in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

si

-0.68

ifi

-0.68

iott

-0.68

mi

-0.67

 Benz

-0.67

omet

-0.66

peria

-0.64

eva

-0.63

mp

-0.60

efer

-0.60

POSITIVE LOGITS

HAM

0.80

tons

0.79

redients

0.79

 conventions

0.77

GGGGGGGG

0.75

enance

0.73

anship

0.71

eering

0.70

anca

0.68

 techniques

0.67

Activations Density 0.072%

phrases related to proportion and quantity

terms related to scaling and adjustments in various contexts

No Comments

No Known Activations