INDEX

Explanations

words related to locations or places

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "ore."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ilts

-0.72

arb

-0.69

iating

-0.66

-+-+

-0.66

urers

-0.64

uation

-0.64

»Ĵ

-0.62

uating

-0.61

 contradictory

-0.60

ENCY

-0.60

POSITIVE LOGITS

tto

1.57

byss

1.30

lli

1.30

tsky

1.29

nz

1.28

tta

1.28

tti

1.26

gon

1.26

ttes

1.20

llo

1.18

Activations Density 0.080%

words related to locations or places

occurrences of the substring "ore."

No Comments

No Known Activations