INDEX

Explanations

financial and economic terms and figures

oai_token-act-pair · gpt-3.5-turbo

punctuation and symbols that indicate lists or negative aspects

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

erest

-0.61

aimon

-0.61

icable

-0.59

ruck

-0.58

 mall

-0.58

 surround

-0.58

pei

-0.58

ji

-0.58

ousel

-0.57

ior

-0.57

POSITIVE LOGITS

something

1.22

including

1.16

albeit

1.06

which

1.05

particularly

1.05

possibly

1.04

meaning

1.04

despite

1.01

especially

1.01

perhaps

1.00

Activations Density 0.175%

financial and economic terms and figures

punctuation and symbols that indicate lists or negative aspects

No Comments

No Known Activations