INDEX

Explanations

mentions of academic degrees, specifically PhDs

oai_token-act-pair · gpt-3.5-turbo

references to doctoral degrees and related academic terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

grounds

-0.83

Peb

-0.73

ality

-0.72

ground

-0.70

ãĥĪ

-0.69

ãĥ¼ãĥ³

-0.69

 Gemini

-0.68

 Samoa

-0.66

theless

-0.66

oldemort

-0.65

POSITIVE LOGITS

 dissertation

1.08

 thesis

0.94

 doctoral

0.93

ĨĴ

0.93

PhD

0.87

essor

0.87

rint

0.85

ysis

0.84

uates

0.82

inct

0.82

Activations Density 0.007%

mentions of academic degrees, specifically PhDs

references to doctoral degrees and related academic terminology

No Comments

No Known Activations