INDEX

Explanations

phrases indicating willingness or suggestion to attempt something

oai_token-act-pair · gpt-3.5-turbo

phrases related to trying or experiencing something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 objects

-0.72

hemat

-0.71

 assemblies

-0.71

 diagrams

-0.70

 branches

-0.70

 mathemat

-0.68

 timelines

-0.67

 Principles

-0.66

 graphs

-0.66

objects

-0.65

POSITIVE LOGITS

TRY

0.82

joy

0.81

kick

0.69

hole

0.68

ibi

0.68

 apiece

0.68

coat

0.67

pour

0.67

iatus

0.67

bike

0.67

Activations Density 0.129%

phrases indicating willingness or suggestion to attempt something

phrases related to trying or experiencing something

No Comments

No Known Activations