INDEX

Explanations

numerical figures related to locations or events, often involving a "Back to Gallery" caption

oai_token-act-pair · gpt-3.5-turbo

numerical data points or statistics related to specific incidents or events

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

DOS

-0.69

assic

-0.63

Cas

-0.63

cas

-0.62

ocre

-0.62

Args

-0.62

anism

-0.62

rats

-0.61

>>>

-0.61

tip

-0.61

POSITIVE LOGITS

 Caption

1.07

 Back

0.89

actionDate

0.72

 Duration

0.70

 Submit

0.70

 BACK

0.69

 NEXT

0.68

 LEVEL

0.68

 Related

0.67

 Sodium

0.67

Activations Density 0.032%

numerical figures related to locations or events, often involving a "Back to Gallery" caption

numerical data points or statistics related to specific incidents or events

No Comments

No Known Activations