INDEX

Explanations

instances where actions lead to significant changes or impacts

oai_token-act-pair · gpt-3.5-turbo

words related to the concept of bringing or causing results or changes

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

livious

-0.79

raid

-0.72

uary

-0.70

debian

-0.69

schild

-0.69

taker

-0.68

dating

-0.63

imon

-0.62

itect

-0.61

 Examiner

-0.61

POSITIVE LOGITS

 forth

1.39

 together

1.02

 forward

0.93

 down

0.88

 smiles

0.83

 attention

0.83

up

0.82

 back

0.81

 misfortune

0.80

 discredit

0.77

Activations Density 0.044%

instances where actions lead to significant changes or impacts

words related to the concept of bringing or causing results or changes

No Comments

No Known Activations