INDEX

Explanations

professional titles or roles

oai_token-act-pair · gpt-3.5-turbo

terms related to professional roles and positions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ļéĨĴ

-0.76

©¶æ

-0.65

 swat

-0.61

 Dreams

-0.60

ãĥ¼ãĥ³

-0.60

ļé

-0.60

PUT

-0.59

 Verb

-0.59

Closure

-0.58

Ħ¢

-0.58

POSITIVE LOGITS

hip

2.10

hips

2.01

paces

1.15

pace

1.09

1.06

peak

1.02

 whom

1.01

who

0.99

who

0.96

doms

0.95

Activations Density 0.477%

professional titles or roles

terms related to professional roles and positions

No Comments

No Known Activations