INDEX

Explanations

differences between two related entities or concepts

oai_token-act-pair · gpt-3.5-turbo

references to comparisons or relationships between two entities

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

RAL

-0.73

 Absent

-0.71

vich

-0.69

gow

-0.67

xus

-0.67

justice

-0.66

aqu

-0.65

%%%%

-0.65

Draft

-0.65

daq

-0.64

POSITIVE LOGITS

 sexes

1.07

 halves

0.98

 extremes

0.91

 poles

0.76

 genders

0.76

 sides

0.76

 coasts

0.76

 continents

0.75

 eras

0.74

 oppos

0.73

Activations Density 0.080%

differences between two related entities or concepts

references to comparisons or relationships between two entities

No Comments

No Known Activations