INDEX

Explanations

references to reflections or metaphorical mirrors in phrases

oai_token-act-pair · gpt-3.5-turbo

references to mirrors and their symbolic significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

cture

-0.75

Ô

-0.72

estern

-0.72

xit

-0.71

erest

-0.69

FE

-0.69

OTT

-0.68

RAFT

-0.67

Northern

-0.67

 Reserve

-0.66

POSITIVE LOGITS

 mirror

1.28

 mirrors

1.19

 mirrored

0.95

 reflection

0.93

 shine

0.85

 Mirror

0.83

 pane

0.79

 reflections

0.78

 symmetry

0.77

 cleaner

0.77

Activations Density 0.006%

references to reflections or metaphorical mirrors in phrases

references to mirrors and their symbolic significance

No Comments

No Known Activations