INDEX

Explanations

references to the name "Jack" with different numerical values for relevance

oai_token-act-pair · gpt-3.5-turbo

the name "Jack" 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

 revolving

-0.69

 laureate

-0.68

pmwiki

-0.68

ndra

-0.67

 describ

-0.66

cation

-0.65

PDATE

-0.64

iferation

-0.64

 gearing

-0.64

 sugg

-0.63

POSITIVE LOGITS

 Sparrow

1.04

intosh

0.92

 Straw

0.91

 Daniels

0.91

knife

0.84

atta

0.82

pots

0.81

hammer

0.81

 Nicholson

0.80

fruit

0.79

Activations Density 0.012%

references to the name "Jack" with different numerical values for relevance

the name "Jack" in various contexts

No Comments

No Known Activations