INDEX

Explanations

quotes or reported speech

oai_token-act-pair · gpt-3.5-turbo

instances of the word "told" used in spoken quotations

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

racted

-0.73

untled

-0.71

immune

-0.68

 ILCS

-0.68

spir

-0.68

reci

-0.67

posure

-0.63

definition

-0.62

icons

-0.59

ashtra

-0.58

POSITIVE LOGITS

tale

0.83

us

0.82

ingly

0.81

him

0.81

me

0.81

 reporters

0.77

IPS

0.76

wered

0.73

tell

0.71

rium

0.71

Activations Density 0.049%

quotes or reported speech

instances of the word "told" used in spoken quotations

No Comments

No Known Activations