INDEX

Explanations

phrases related to movie titles, names, and random alphanumeric sequences

oai_token-act-pair · gpt-3.5-turbo

titles or names related to various media or entertainment formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ritical

-0.59

inventoryQuantity

-0.55

otally

-0.54

meal

-0.54

canon

-0.54

aeper

-0.53

astical

-0.53

 flexibility

-0.53

\":

-0.53

 loopholes

-0.53

POSITIVE LOGITS

LLP

0.99

ensis

0.98

Ltd

0.95

!.

0.87

?,

0.82

 fame

0.82

.;

0.81

Jr

0.79

!,

0.78

/,

0.77

Activations Density 0.561%

phrases related to movie titles, names, and random alphanumeric sequences

titles or names related to various media or entertainment formats

No Comments

No Known Activations