INDEX

Explanations

words related to changing one's mind

oai_token-act-pair · gpt-3.5-turbo

mentions of "mind" and "minds."

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

EStreamFrame

-0.69

Pie

-0.68

é¾

-0.66

equ

-0.65

ammy

-0.64

 Municip

-0.63

 Underworld

-0.60

rylic

-0.60

heny

-0.60

EPS

-0.59

POSITIVE LOGITS

fulness

1.03

ets

1.00

storms

0.94

locks

0.85

ulsion

0.80

endment

0.79

lessly

0.78

eting

0.75

lock

0.73

 wand

0.72

Activations Density 0.016%

words related to changing one's mind

mentions of "mind" and "minds."

No Comments

No Known Activations