INDEX

Explanations

mentions of the term "ore"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "ore"

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uation

-0.70

ilts

-0.69

uating

-0.69

keye

-0.66

-+-+

-0.66

arb

-0.66

imov

-0.65

urers

-0.65

onomic

-0.65

uate

-0.62

POSITIVE LOGITS

tto

1.39

byss

1.24

gon

1.19

tsky

1.18

lli

1.18

nz

1.13

ttes

1.12

tti

1.11

xia

1.11

tta

1.11

Activations Density 0.053%

mentions of the term "ore"

occurrences of the word "ore"

No Comments

No Known Activations