INDEX

Explanations

instances where a situation is described with a reasonable or logical conditional comparison

oai_token-act-pair · gpt-3.5-turbo

conditional phrases that express hypothetical situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

verts

-0.82

vez

-0.72

akia

-0.69

hiba

-0.67

washer

-0.67

vert

-0.65

ethy

-0.65

eph

-0.63

phrine

-0.61

arry

-0.61

POSITIVE LOGITS

 nothing

0.80

 TAMADRA

0.80

yip

0.79

 they

0.71

 possessed

0.67

 menacing

0.66

 Garc

0.65

 incapable

0.64

 THEY

0.62

inventoryQuantity

0.60

Activations Density 0.038%

instances where a situation is described with a reasonable or logical conditional comparison

conditional phrases that express hypothetical situations

No Comments

No Known Activations