INDEX

Explanations

words related to political or government actions

oai_token-act-pair · gpt-3.5-turbo

phrases indicating levels or scales of action and accountability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Interested

-0.70

hin

-0.68

photo

-0.64

NESS

-0.61

Reply

-0.60

fill

-0.60

gins

-0.60

idon

-0.59

seeing

-0.59

[+

-0.58

POSITIVE LOGITS

 least

1.12

onement

1.11

 stake

1.05

 conferences

0.93

 universities

0.91

 airports

0.89

abase

0.89

 home

0.88

omics

0.86

 logger

0.82

Activations Density 0.125%

words related to political or government actions

phrases indicating levels or scales of action and accountability

No Comments

No Known Activations