INDEX

Explanations

verbs or phrases related to making decisions or conclusions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "determine" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Lens

-0.68

GB

-0.63

far

-0.62

exper

-0.61

door

-0.60

hearted

-0.59

nic

-0.59

science

-0.58

 Zeal

-0.58

ita

-0.57

POSITIVE LOGITS

 determine

1.00

cules

0.91

 determines

0.90

ially

0.85

ministic

0.81

ively

0.78

uate

0.75

 evaluates

0.74

initions

0.73

Ļ½

0.73

Activations Density 0.011%

verbs or phrases related to making decisions or conclusions

instances of the word "determine" in various contexts

No Comments

No Known Activations