INDEX

Explanations

nested lists or nested list-related operations

oai_token-act-pair · gpt-3.5-turbo

references to lists in programming contexts

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

icago

-0.76

asio

-0.73

aneous

-0.64

 streng

-0.63

irgin

-0.62

 charism

-0.62

 elbows

-0.62

ione

-0.59

 whist

-0.58

undai

-0.58

POSITIVE LOGITS

erv

1.09

cloth

0.89

ener

0.84

ucc

0.82

view

0.81

room

0.81

icles

0.80

erve

0.79

 comprehens

0.79

mates

0.79

Activations Density 0.030%

nested lists or nested list-related operations

references to lists in programming contexts

No Comments

No Known Activations