INDEX

Explanations

text related to guidance or being led in various contexts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "guided" and related terms

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

ellen

-0.71

skirts

-0.67

 ante

-0.67

LAN

-0.65

Fax

-0.64

ylum

-0.63

eon

-0.62

 Lyons

-0.61

enos

-0.61

ouf

-0.59

POSITIVE LOGITS

 guided

0.97

 guiding

0.93

itudinal

0.85

 guides

0.84

 tours

0.83

Ń·

0.78

 meditation

0.77

zag

0.77

htaking

0.75

 steered

0.75

Activations Density 0.019%

text related to guidance or being led in various contexts

instances of the word "guided" and related terms

No Comments

No Known Activations