INDEX

Explanations

the name "Carol" or "Daryl" in the text

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name "Carol" 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

Reply

-0.73

asonable

-0.71

claimer

-0.71

fix

-0.68

utils

-0.68

facts

-0.68

hower

-0.67

election

-0.67

flat

-0.67

change

-0.67

POSITIVE LOGITS

 Burnett

1.02

otte

0.95

inas

0.88

 Carol

0.81

Sue

0.79

 McDonnell

0.77

ucci

0.76

lee

0.74

olina

0.74

Rae

0.73

Activations Density 0.014%

the name "Carol" or "Daryl" in the text

occurrences of the name "Carol" in various contexts

No Comments

No Known Activations