INDEX

Explanations

the word "one" followed by a number, potentially indicating a specific entity or concept

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "one" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ooks

-0.73

ories

-0.69

ãģĤ

-0.64

ãĤµ

-0.64

inders

-0.63

 actionGroup

-0.60

ãĥ©ãĥ³

-0.60

lov

-0.59

folk

-0.59

ãĤ½

-0.59

POSITIVE LOGITS

 hundred

0.96

 dimensional

0.88

 Hundred

0.88

rency

0.88

 sided

0.88

esan

0.80

 thousand

0.78

 embodiment

0.77

 thing

0.76

dimensional

0.74

Activations Density 0.154%

the word "one" followed by a number, potentially indicating a specific entity or concept

occurrences of the word "one" in various contexts

No Comments

No Known Activations