INDEX

Explanations

proper nouns, specifically names of individuals in news reports or incidents

oai_token-act-pair · gpt-3.5-turbo

instances of the preposition "of"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 functioning

-0.77

attribute

-0.72

 disse

-0.72

 evaluates

-0.68

 derog

-0.68

 dictate

-0.68

 explan

-0.66

 vulner

-0.66

pard

-0.66

 disag

-0.65

POSITIVE LOGITS

 Anaheim

1.00

 Lancaster

0.98

 Queens

0.98

 Stamford

0.97

 Bethlehem

0.95

 Wilmington

0.94

 Calgary

0.94

 Rochester

0.94

 Omaha

0.93

 Auckland

0.93

Activations Density 0.070%

proper nouns, specifically names of individuals in news reports or incidents

instances of the preposition "of"

No Comments

No Known Activations