INDEX

Explanations

references to the year 2017

oai_token-act-pair · gpt-3.5-turbo

occurrences of the year 2017 in various contexts

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

 repro

-0.66

cold

-0.64

 Mell

-0.62

tub

-0.62

Bone

-0.61

ym

-0.60

imore

-0.59

 snipp

-0.59

 Cold

-0.58

cipl

-0.58

POSITIVE LOGITS

halla

0.83

âĢ²

0.69

Marginal

0.68

âĢ³

0.68

âĢķ

0.66

 Winners

0.66

ership

0.65

å¹

0.64

ESV

0.64

Activations Density 0.037%

references to the year 2017

occurrences of the year 2017 in various contexts

No Comments

No Known Activations