INDEX

Explanations

the word "then" followed by a numerical value

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

toe

-0.74

md

-0.61

illon

-0.61

 Parables

-0.60

 Shore

-0.59

acts

-0.59

des

-0.57

 Peninsula

-0.57

 triangles

-0.56

ting

-0.54

POSITIVE LOGITS

-'

0.90

-,

0.77

 Yugoslav

0.74

0.73

 proceeded

0.72

Reviewer

0.71

arily

0.67

-.

0.66

ose

0.65

ally

0.65

Activations Density 0.039%

the word "then" followed by a numerical value

No Comments

No Known Activations