INDEX

Explanations

individual instances or mentions of the word "solo" in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to solo performances or solo activities

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

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

 Hussain

-0.73

tics

-0.68

older

-0.66

cale

-0.66

nah

-0.65

 GOODMAN

-0.64

imaru

-0.63

truth

-0.63

 Glover

-0.61

generic

-0.61

POSITIVE LOGITS

iliary

0.92

spective

0.84

ists

0.84

spection

0.83

istic

0.82

ist

0.82

 piano

0.80

 ventures

0.77

ism

0.76

 acoustic

0.76

Activations Density 0.036%

individual instances or mentions of the word "solo" in various contexts

references to solo performances or solo activities

No Comments

No Known Activations