INDEX

Explanations

specific numbers related to locations or details

oai_token-act-pair · gpt-3.5-turbo

numerical identifiers, often referencing specific locations or rooms

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

pects

-0.65

iques

-0.63

asts

-0.57

soever

-0.56

 suggestions

-0.55

tones

-0.55

 opin

-0.54

 warranties

-0.54

nom

-0.54

orians

-0.53

POSITIVE LOGITS

1.10

1.08

1.07

1.06

1.05

Activations Density 0.112%

specific numbers related to locations or details

numerical identifiers, often referencing specific locations or rooms

No Comments

No Known Activations