INDEX

Explanations

numeric coordinates or locations mentioned in text

oai_token-act-pair · gpt-3.5-turbo

terms related to coordinates and coordination

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

anol

-0.75

culosis

-0.74

lasses

-0.73

stakes

-0.73

tle

-0.73

shared

-0.71

aneers

-0.70

trap

-0.69

spl

-0.69

ħĭ

-0.69

POSITIVE LOGITS

inates

1.35

 coordinates

1.21

 coordinate

0.94

inators

0.87

inated

0.87

 Coord

0.85

ially

0.83

 coord

0.82

inate

0.78

coord

0.75

Activations Density 0.012%

numeric coordinates or locations mentioned in text

terms related to coordinates and coordination

No Comments

No Known Activations