INDEX

Explanations

comparisons of position or progress

oai_token-act-pair · gpt-3.5-turbo

phrases indicating relative position or status, particularly in relation to being behind, off, ahead, or out of place

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

cu

-0.88

GM

-0.70

Ca

-0.70

©¶æ

-0.70

du

-0.68

Gi

-0.68

DC

-0.67

ha

-0.67

uff

-0.66

Chem

-0.66

POSITIVE LOGITS

sighted

1.13

 academ

0.84

 bounds

0.71

enance

0.70

reaching

0.69

fetched

0.69

erous

0.68

 paradise

0.68

 reach

0.67

shore

0.66

Activations Density 0.114%

comparisons of position or progress

phrases indicating relative position or status, particularly in relation to being behind, off, ahead, or out of place

No Comments

No Known Activations