INDEX

Explanations

numbers related to quantities or statistics

oai_token-act-pair · gpt-3.5-turbo

phrases related to numerical data or statistics

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

enment

-0.81

 Deliver

-0.73

Ļ½

-0.68

Interstitial

-0.68

 coli

-0.67

CLA

-0.64

Ai

-0.62

entimes

-0.62

route

-0.61

Dialogue

-0.61

POSITIVE LOGITS

 balloons

0.73

 brackets

0.69

 inflated

0.69

 graphs

0.68

 underest

0.65

é¾įå

0.65

emark

0.65

 sums

0.64

oola

0.64

 snapshots

0.63

Activations Density 0.394%

numbers related to quantities or statistics

phrases related to numerical data or statistics

No Comments

No Known Activations