INDEX

Explanations

adjectives describing negative or unpleasant characteristics or situations

oai_token-act-pair · gpt-3.5-turbo

descriptors related to negative or severe situations

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

afort

-0.73

 actionGroup

-0.71

————

-0.68

adesh

-0.67

 enzyme

-0.65

Thumbnail

-0.65

 congr

-0.63

ertodd

-0.62

 granted

-0.62

 streng

-0.60

POSITIVE LOGITS

oire

1.19

dark

1.18

ly

1.04

aces

0.97

ace

0.97

acing

0.92

lock

0.91

grim

0.90

 prog

0.88

 outlook

0.87

Activations Density 0.073%

adjectives describing negative or unpleasant characteristics or situations

descriptors related to negative or severe situations

No Comments

No Known Activations