INDEX

Explanations

years mentioned in historical contexts

oai_token-act-pair · gpt-3.5-turbo

mentions of decades, specifically the 1950s to 1990s

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

edIn

-0.85

ournals

-0.80

 snipp

-0.67

Maker

-0.65

 slips

-0.63

acts

-0.63

inventoryQuantity

-0.62

 Strongh

-0.62

bind

-0.60

Planet

-0.60

POSITIVE LOGITS

 onwards

1.15

 onward

1.06

era

0.92

ixties

0.85

Era

0.85

 nostalg

0.80

-'

0.75

olutions

0.72

hey

0.71

BCE

0.71

Activations Density 0.053%

years mentioned in historical contexts

mentions of decades, specifically the 1950s to 1990s

No Comments

No Known Activations