INDEX

Explanations

information related to achievements, awards, and recognition

oai_token-act-pair · gpt-3.5-turbo

references to achievement and recognition

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

vous

-0.81

 contrace

-0.79

Plan

-0.71

addons

-0.70

glers

-0.69

odcast

-0.67

dayName

-0.67

onday

-0.65

 Tackle

-0.65

onics

-0.65

POSITIVE LOGITS

 accol

1.87

 notoriety

1.76

 fame

1.74

 acclaim

1.71

 renown

1.58

 admiration

1.49

 awards

1.48

 reputation

1.46

 recognition

1.44

 honors

1.41

Activations Density 0.614%

information related to achievements, awards, and recognition

references to achievement and recognition

No Comments

No Known Activations