INDEX

Explanations

instances of explosions

oai_token-act-pair · gpt-3.5-turbo

references to explosions and their impacts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

nda

-0.83

tern

-0.70

lege

-0.70

stra

-0.69

 dearly

-0.66

ndra

-0.66

ippi

-0.65

leg

-0.64

iff

-0.64

dden

-0.64

POSITIVE LOGITS

 explosion

0.94

 furnace

0.84

 explosions

0.82

 bursting

0.82

 furn

0.82

 radius

0.80

opl

0.80

hower

0.79

 burst

0.77

 Explosion

0.76

Activations Density 0.019%

instances of explosions

references to explosions and their impacts

No Comments

No Known Activations