INDEX

Explanations

words related to making something visible or obvious

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "manifest" in various forms

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

apter

-0.84

zanne

-0.75

pei

-0.74

restling

-0.71

ERA

-0.69

umph

-0.69

zzi

-0.67

maxwell

-0.66

awks

-0.66

pg

-0.65

POSITIVE LOGITS

 manifests

0.93

 manifestation

0.91

 manifestations

0.89

 destiny

0.87

eering

0.85

 manifest

0.85

iary

0.81

ly

0.80

oes

0.77

antly

0.75

Activations Density 0.017%

words related to making something visible or obvious

occurrences of the word "manifest" in various forms

No Comments

No Known Activations