INDEX

Explanations

locations and dates

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "in" and related temporal indicators

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

aters

-0.67

redo

-0.66

win

-0.65

anamo

-0.64

heet

-0.61

emo

-0.61

Others

-0.60

phans

-0.60

aea

-0.60

cia

-0.60

POSITIVE LOGITS

 Week

0.77

 winter

0.76

 July

0.74

 August

0.74

 December

0.74

 January

0.73

 March

0.73

 February

0.73

 June

0.72

Activations Density 0.290%

locations and dates

occurrences of the word "in" and related temporal indicators

No Comments

No Known Activations