INDEX

Explanations

words related to experimentation and trying out new things

oai_token-act-pair · gpt-3.5-turbo

instances of experimentation and related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Cheong

-0.72

cube

-0.69

Heb

-0.67

translation

-0.67

mary

-0.65

 Supporting

-0.65

lat

-0.64

die

-0.64

ens

-0.63

trans

-0.63

POSITIVE LOGITS

 experimenting

1.10

 experimented

0.98

 tink

0.96

 withd

0.96

 experimentation

0.95

odox

0.90

redients

0.89

quished

0.87

GGGGGGGG

0.85

iences

0.74

Activations Density 0.014%

words related to experimentation and trying out new things

instances of experimentation and related concepts

No Comments

No Known Activations