INDEX

Explanations

time-related expressions, specifically durations

oai_token-act-pair · gpt-3.5-turbo

references to time durations, particularly phrases indicating "one" or "two" years or "one and a half" years

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ãĥ¤

-0.71

alg

-0.71

Sin

-0.65

Screen

-0.64

silver

-0.61

 Sins

-0.61

enburg

-0.60

 filib

-0.60

individual

-0.58

 Cron

-0.58

POSITIVE LOGITS

ago

0.93

depending

0.84

 depending

0.81

 respectively

0.77

 worth

0.77

 glance

0.73

 glimpse

0.72

 before

0.71

 apiece

0.69

 fueling

0.68

Activations Density 0.050%

time-related expressions, specifically durations

references to time durations, particularly phrases indicating "one" or "two" years or "one and a half" years

No Comments

No Known Activations