INDEX

Explanations

words related to people's names or references to organizations

oai_token-act-pair · gpt-3.5-turbo

occurrences of the segment "pe"

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

é¾

-0.93

 GOODMAN

-0.74

 microscope

-0.68

 swirling

-0.63

 Kirin

-0.63

 simultane

-0.62

enegger

-0.62

 nets

-0.62

 calculus

-0.60

ãĤ¨ãĥ«

-0.60

POSITIVE LOGITS

cific

1.10

anut

1.03

oples

1.00

uberty

0.99

asant

0.97

ller

0.97

rer

0.96

cies

0.92

rers

0.90

cial

0.88

Activations Density 0.012%

words related to people's names or references to organizations

occurrences of the segment "pe"

No Comments

No Known Activations