INDEX

Explanations

phrases referring to different types of objects or things

oai_token-act-pair · gpt-3.5-turbo

phrases that express various kinds of relationships or associations

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

seless

-0.82

 Seconds

-0.79

ysc

-0.76

enes

-0.76

essors

-0.76

iest

-0.75

erest

-0.74

apest

-0.72

ores

-0.71

 Sands

-0.71

POSITIVE LOGITS

 imaginable

0.79

 kindred

0.77

 meaningful

0.76

 harassment

0.75

 crossover

0.74

 sudden

0.73

 intermediary

0.72

 validation

0.72

 foothold

0.70

 adversity

0.69

Activations Density 0.057%

phrases referring to different types of objects or things

phrases that express various kinds of relationships or associations

No Comments

No Known Activations