INDEX

Explanations

words related to mythical or fictional creatures, particularly monsters

oai_token-act-pair · gpt-3.5-turbo

references to "monsters" 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

arers

-0.82

ories

-0.82

atche

-0.77

omers

-0.77

blance

-0.75

 cort

-0.75

lar

-0.74

lav

-0.74

lay

-0.72

nor

-0.72

POSITIVE LOGITS

 monster

1.14

 monsters

1.09

 beasts

0.92

 Fenrir

0.86

 beast

0.85

 Monstrous

0.80

 Beasts

0.79

monster

0.78

 Monsters

0.77

Monster

0.77

Activations Density 0.014%

words related to mythical or fictional creatures, particularly monsters

references to "monsters" in various contexts

No Comments

No Known Activations